Load scene sequence generation method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a load scene sequence generation method and device, computer equipment, a readable storage medium and a program product. The method comprises the following steps: dividing the annual total load sequence of a target region to obtain an air temperature factor sensitive load sequence and a non-air temperature factor sensitive load sequence; obtaining a temperature change model based on a daily target temperature corresponding to each first daily load sequence in the temperature factor sensitive load sequence and a daily maximum temperature change load corresponding to each first daily load sequence; generating a first load scene sequence according to the temperature change model; counting a plurality of different characteristic indexes of each second-day load sequence in the non-air-temperature-factor sensitive load sequence to obtain a joint probability distribution model; and generating a second load scene sequence according to the joint probability distribution model. By adopting the method, the accuracy of load scene sequence generation can be improved.

Description

Technical Field

[0001] The present application relates to the field of power system control technology, and in particular to a load scenario generation method, device, computer equipment, readable storage medium and program product. Background Art

[0002] The load scenario generation method can effectively describe the uncertainty on both the source and load sides of the power system, and has been widely used in the medium- and long-term planning and optimization models of power systems.

[0003] For the generation of medium- and long-term load scenario sequences, most of the existing studies obtain the load scenario sequence for the target planning year by clustering historical load data, or introduce deep learning methods such as Generative Adversarial Networks (GAN) and Variational AutoEncoder (VAE) to generate load scenario sequences.

[0004] However, the existing research methods need to be improved to further improve the accuracy of load scenario sequence generation and thus improve the reliability of medium- and long-term planning of power systems. Summary of the invention

[0005] Based on this, it is necessary to provide a load scenario generation method, apparatus, computer equipment, readable storage medium and program product that can further improve the accuracy of load scenario sequence generation in response to the above technical problems.

[0006] In a first aspect, the present application provides a method for generating a load scenario sequence, comprising: dividing the total load sequence of the target area throughout the year to obtain a temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence; obtaining a temperature change model based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence; generating a first load scenario sequence according to the temperature change model; counting multiple different characteristic indicators of each second-day load sequence in the non-temperature factor sensitive load sequence to obtain a joint probability distribution model; generating a second load scenario sequence according to the joint probability distribution model.

[0007] In one of the embodiments, before obtaining the temperature change model based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence, the method also includes: for each first-day load sequence in the temperature factor sensitive load sequence, based on a pre-constructed temperature cumulative effect optimization model, correcting the daily temperature corresponding to the first-day load sequence to obtain the daily target temperature corresponding to the first-day load sequence.

[0008] In one embodiment, the temperature change model includes a temperature sensitivity coefficient, and a first load scenario sequence is generated according to the temperature change model, including: correcting the temperature sensitivity coefficient in the temperature change model according to the electrification level of the year to be predicted to obtain a corrected temperature change model; inputting the temperature of the day to be predicted into the corrected temperature change model to obtain the daily maximum temperature change load of the day to be predicted; selecting a first preset number of first-day load sequences from the temperature factor sensitive load sequence according to the daily maximum temperature change load of the day to be predicted; and obtaining a first load scenario sequence according to the selected first preset number of first-day load sequences.

[0009] In one embodiment, a second load scenario sequence is generated according to a joint probability distribution model, including: sampling a plurality of different characteristic indicators of the day to be predicted from the joint probability distribution model; screening a benchmark load sequence from a plurality of second-day load sequences of non-temperature factor sensitive load sequences according to the plurality of different characteristic indicators of the day to be predicted; and obtaining a second load scenario sequence based on a pre-constructed load sequence optimization model and a benchmark load sequence.

[0010] In one of the embodiments, based on multiple different characteristic indicators of the day to be predicted, a benchmark load sequence is screened from multiple second-day load sequences of the non-temperature factor sensitive load sequence, including: obtaining multiple different characteristic indicators of each second-day load sequence, for each second-day load sequence, performing similarity calculations on multiple different characteristic indicators of the second-day load sequence and multiple different characteristic indicators of the day to be predicted to obtain the similarity of the second-day load sequence; and selecting the second-day load sequence with the greatest similarity from the multiple second-day load sequences as the benchmark load sequence.

[0011] In one of the embodiments, the total load sequence of the target area throughout the year is divided to obtain a temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence, including: deleting the daily load sequence corresponding to holidays in the total load sequence to obtain the deleted total load sequence; dividing the deleted total load sequence according to the geographical environment of the target area to obtain an initial temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence; based on the non-temperature factor sensitive load sequence, deleting the non-temperature factor sensitive load contained in the initial temperature factor sensitive load sequence to obtain a temperature factor sensitive load sequence.

[0012] In the second aspect, the present application also provides a load scenario sequence generating device, including: a division module, used to divide the total load sequence of the target area throughout the year to obtain a temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence; a first generation module, used to obtain a temperature change model based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence; generate a first load scenario sequence according to the temperature change model; a second generation module, used to count multiple different characteristic indicators of each second-day load sequence in the non-temperature factor sensitive load sequence to obtain a joint probability distribution model; generate a second load scenario sequence according to the joint probability distribution model.

[0013] In one embodiment, the device also includes a correction module for correcting the daily temperature corresponding to the first-day load sequence in the temperature-sensitive load sequence based on a pre-built temperature cumulative effect optimization model to obtain the daily target temperature corresponding to the first-day load sequence.

[0014] In one embodiment, the temperature change model includes a temperature sensitivity coefficient, and a first generation module is specifically used to correct the temperature sensitivity coefficient in the temperature change model according to the electrification level of the year to be predicted to obtain a corrected temperature change model; the temperature of the day to be predicted is input into the corrected temperature change model to obtain the daily maximum temperature change load of the day to be predicted; according to the daily maximum temperature change load of the day to be predicted, a first preset number of first-day load sequences are selected from the temperature factor sensitive load sequence; based on the selected first preset number of first-day load sequences, a first load scenario sequence is obtained.

[0015] In one of the embodiments, the second generation module is specifically used to sample multiple different characteristic indicators of the day to be predicted from the joint probability distribution model; based on the multiple different characteristic indicators of the day to be predicted, a benchmark load sequence is screened from multiple second-day load sequences of the non-temperature factor sensitive load sequence; based on the pre-constructed load sequence optimization model and the benchmark load sequence, a second load scenario sequence is obtained.

[0016] In one embodiment, the second generation module is specifically used to obtain multiple different characteristic indicators of each second-day load sequence, and for each second-day load sequence, perform similarity calculations on the multiple different characteristic indicators of the second-day load sequence and the multiple different characteristic indicators of the day to be predicted to obtain the similarity of the second-day load sequence; and select the second-day load sequence with the greatest similarity from the multiple second-day load sequences as the benchmark load sequence.

[0017] In one of the embodiments, the division module is specifically used to delete the daily load sequence corresponding to holidays in the total load sequence to obtain the total load sequence after deletion; the total load sequence after deletion is divided according to the geographical environment of the target area to obtain an initial temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence; based on the non-temperature factor sensitive load sequence, the non-temperature factor sensitive load contained in the initial temperature factor sensitive load sequence is deleted to obtain a temperature factor sensitive load sequence.

[0018] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the methods described in the first aspect when executing the computer program.

[0019] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any one of the methods described in the first aspect are implemented.

[0020] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any one of the methods described in the first aspect.

[0021] The above-mentioned load scenario sequence generation method, device, computer equipment, readable storage medium and program product divide the total load sequence of the target area throughout the year to obtain a temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence. Then, based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence, a temperature change model is obtained; and a first load scenario sequence is generated according to the temperature change model. And a plurality of different characteristic indicators of each second-day load sequence in the non-temperature factor sensitive load sequence are statistically analyzed to obtain a joint probability distribution model; and a second load scenario sequence is generated according to the joint probability distribution model. In this way, by dividing the total load sequence of the target area throughout the year to obtain a temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence, the first load scenario sequence and the second load scenario sequence obtained based on the temperature factor sensitive load sequence and the non-temperature factor sensitive load sequence respectively take into account the impact of temperature factors on the load, so that the accuracy of the load scenario sequence obtained is higher, and the reliability of the long-term planning of the power system is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1 is an application environment diagram of a load scenario sequence generation method in an embodiment;

[0024] Figure 2 A schematic diagram of a flow chart of a method for generating a load scenario sequence in one embodiment;

[0025] Figure 3 A schematic diagram of the principle of generating a load scenario sequence in one embodiment;

[0026] Figure 4 A graph showing the year-on-year variation trend of the summer temperature sensitivity coefficients of three provinces and regions in an embodiment;

[0027] Figure 5 A comparison chart of characteristic indicators in an embodiment;

[0028] Figure 6 A probability density curve comparison diagram in one embodiment;

[0029] Figure 7 A comparison diagram of probability density curves of summer daily load peak values ​​under different development scenarios in an embodiment;

[0030] Figure 8 is a comparison diagram of summer load sequences under different development scenarios in an embodiment;

[0031] Fig. 9 A comparison diagram of cumulative distribution curves of online capacity of thermal power units in summer under different load scenarios in an embodiment;

[0032] Fig.10 is a structural block diagram of a load scenario sequence generating device in an embodiment;

[0033] Fig.11 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0035] The proportion of renewable energy and the level of electrification of terminal energy consumption continue to increase, and the transformation of energy structure has brought new problems to the power system. As the substitution of electric energy continues to advance, the proportion of summer cooling load and winter heating load continues to increase, and the winter and summer double peak characteristics of load are becoming more and more obvious. Many regions have experienced power shortages under continuous high temperature or cold wave weather. Therefore, it is necessary to consider the impact of the increase in electrification level and the uncertainty of temperature factors on the load side in power system planning.

[0036] The load scenario generation method can effectively describe the uncertainty of both the source and load sides of the power system, and has been widely used in the medium- and long-term planning and optimization models of the power system. For the generation of medium- and long-term load scenarios, most existing studies first obtain multiple representative typical daily or annual load sequences from historical load data through manual screening or clustering methods, and then scale the historical typical load scenarios based on the medium- and long-term electricity or load peak forecast results to obtain the load scenario sequence for the target planning year. There are also studies that introduce deep learning methods such as generative adversarial networks and variational autoencoders to generate load scenario sequences.

[0037] The evaluation of the above load scenario generation methods is usually based on probability distribution characteristics and time series characteristics related indicators to measure the similarity between the generated limited number of scenarios and historical load data. However, the main problem in the generation of load scenarios applied to medium- and long-term planning and operation analysis of power systems is that the load scenario sequence generated by modeling based on limited historical data characteristics can only reproduce the historical load characteristics. The goal of load scenario generation is still similar to historical loads. It is difficult to fully model the impact of electrification levels and extreme weather events such as continuous high temperatures / cold waves on power loads, and it is difficult to fully characterize the changes and development trends of future power loads for medium- and long-term power system planning.

[0038] Therefore, the existing research methods need to be improved to further improve the accuracy of load scenario sequence generation, thereby improving the reliability of medium- and long-term planning of power systems.

[0039] The load scenario sequence generation method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, tablet computers, and mobile phones. The server 104 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0040] In an exemplary embodiment, Figure 2 As shown, a load scenario sequence generation method is provided, and the method is applied to Figure 1 The terminal in is taken as an example to illustrate, including the following steps 201 to 203. Among them:

[0041] Step 201 , dividing the total load sequence of the target area throughout the year to obtain a temperature-sensitive load sequence and a non-temperature-sensitive load sequence.

[0042] Among them, the target area can be a region divided according to climate, such as the target area can be the northeast region, the southern region or other regions except the northeast region and the southern region; the target area can also be a region divided according to province, such as the target area can be a certain province.

[0043] The total load sequence of the target area for the whole year includes the daily load sequence corresponding to each day of the target area in a year. The daily load sequence is a sequence consisting of the loads corresponding to 24 time points in a day. Therefore, the total load sequence of the target area for the whole year can be regarded as a 365×24 load matrix. Of course, for the total load sequence of the target area for many years in history, it can be regarded as multiple 365×24 load matrices.

[0044] Temperature sensitive load sequence refers to the load sequence of loads affected by temperature. Non-temperature sensitive load sequence refers to the load sequence of loads not affected by temperature. Loads not affected by temperature are equivalent to loads with relatively stable values ​​over a period of time.

[0045] Optionally, select multiple different characteristic indicators to perform characteristic analysis on the daily load sequence in the total load sequence to obtain the characteristic analysis results of each daily load sequence; select multiple different meteorological factors to perform correlation analysis on the daily load sequence in the total load sequence to obtain the correlation analysis results of each daily load sequence and different meteorological factors. According to the characteristic analysis results of each daily load sequence and the correlation analysis results of each daily load sequence and different meteorological factors, the daily load sequence is divided to obtain the initial temperature factor sensitive load sequence and the non-temperature factor sensitive load sequence. In other words, the daily load sequence in the total load sequence is divided according to the geographical environment of the target area.

[0046] In one example, characteristic indicators such as daily load mean, peak value, peak-to-valley difference, peak load rate and valley load rate are selected to perform characteristic analysis on each daily load sequence, and characteristic analysis results such as the volatility of daily load and the high peak level of daily load corresponding to each daily load sequence are obtained. The correlation coefficients between meteorological factors such as daily maximum temperature, average temperature, minimum temperature, daily average wind speed (wind chill effect) and daily load corresponding to each daily load sequence are calculated. Based on the characteristic analysis results such as the volatility of daily load and the high peak level of daily load corresponding to each daily load sequence, as well as the correlation coefficients between different meteorological factors and the daily load corresponding to each daily load sequence, each daily load sequence is divided to obtain the initial temperature factor sensitive load sequence and the non-temperature factor sensitive load sequence.

[0047] For example, based on the characteristic analysis results of each daily load sequence and the correlation analysis results between each daily load sequence and different meteorological factors, the daily load sequences corresponding to June, July, August, December, January and February of the year are divided into load sequences sensitive to initial temperature factors, and the daily load sequences corresponding to March, April, May, September, October and November are divided into load sequences sensitive to non-temperature factors.

[0048] After obtaining the initial temperature factor sensitive load sequence and the non-temperature factor sensitive load sequence, since the initial temperature factor sensitive load sequence includes the non-temperature factor sensitive load, and the non-temperature factor sensitive load sequence represents the non-temperature factor sensitive load, therefore, based on the non-temperature factor sensitive load sequence, the non-temperature factor sensitive load contained in the initial temperature factor sensitive load sequence can be deleted to obtain the temperature factor sensitive load sequence.

[0049] Step 202, based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence, a temperature change model is obtained; and a first load scenario sequence is generated according to the temperature change model.

[0050] Among them, in order to facilitate the distinction between the daily load sequence in the temperature factor sensitive load sequence and the daily load sequence in the non-temperature factor sensitive load sequence, the daily load sequence in the temperature factor sensitive load sequence is regarded as the first day load sequence, and the daily load sequence in the non-temperature factor sensitive load sequence is regarded as the second day load sequence.

[0051] The daily maximum temperature change load corresponding to the first day's load sequence refers to the peak value in the first day's load sequence. The daily target temperature can be the daily average temperature corresponding to the first day's load sequence; it can also be the corrected daily average temperature obtained by correcting the daily average temperature corresponding to the first day's load sequence considering the cumulative effect of temperature.

[0052] Optionally, a multivariate linear regression model is used to fit the daily target temperatures corresponding to multiple first-day load sequences and the daily maximum temperature change loads corresponding to multiple first-day load sequences to obtain a temperature change model.

[0053] In one example, a first load scenario sequence is generated according to a temperature change model, including: inputting the temperature of the day to be predicted into a modified temperature change model to obtain the daily maximum temperature change load of the day to be predicted; selecting a first preset number of first-day load sequences from a temperature factor sensitive load sequence according to the daily maximum temperature change load of the day to be predicted; and obtaining a first load scenario sequence according to the selected first preset number of first-day load sequences.

[0054] In another example, generating a first load scenario sequence according to a temperature change model includes: correcting a temperature sensitivity coefficient in the temperature change model according to an electrification level in a predicted year to obtain a corrected temperature change model; and generating the first load scenario sequence according to the corrected temperature change model.

[0055] Step 203 , counting multiple different characteristic indicators of each second-day load sequence in the non-temperature factor sensitive load sequence to obtain a joint probability distribution model; and generating a second load scenario sequence according to the joint probability distribution model.

[0056] The multiple different characteristic indicators may be at least two of the daily load mean, peak value, peak-to-valley difference, peak load rate, and valley load rate, etc. Preferably, the multiple different characteristic indicators are the daily load mean, peak value, peak-to-valley difference, peak load rate, and valley load rate.

[0057] Optionally, multiple different characteristic indicators of each second-day load sequence in the non-temperature factor sensitive load sequence are counted; for each characteristic indicator, a non-parametric kernel density estimation method is used to fit the characteristic indicators targeted by multiple second-day load sequences to obtain the marginal probability distribution of the targeted characteristic indicators; considering the correlation between the characteristic indicators, a Vine-Copula function is used to construct a joint probability distribution model based on the marginal probability distribution of multiple characteristic indicators, from which multiple different characteristic indicators with one-to-one correspondence can be extracted.

[0058] In one example, a second load scenario sequence is generated according to a joint probability distribution model, including: sampling a plurality of different characteristic indicators of the day to be predicted from the joint probability distribution model; screening a benchmark load sequence from a plurality of second-day load sequences of non-temperature factor sensitive load sequences according to the plurality of different characteristic indicators of the day to be predicted; and obtaining a second load scenario sequence based on a pre-constructed load sequence optimization model and a benchmark load sequence.

[0059] The above-mentioned load scenario sequence generation method divides the total load sequence of the target area throughout the year to obtain a temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence. Then, based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence, a temperature change model is obtained; and the first load scenario sequence is generated according to the temperature change model. And a plurality of different characteristic indicators of each second-day load sequence in the non-temperature factor sensitive load sequence are statistically analyzed to obtain a joint probability distribution model, and the joint probability distribution model includes a probability distribution model corresponding to each characteristic indicator; and the second load scenario sequence is generated according to the joint probability distribution model. In this way, the temperature factor sensitive load sequence and the non-temperature factor sensitive load sequence are obtained by dividing the total load sequence of the target area throughout the year, thereby obtaining a first load scenario sequence and a second load scenario sequence based on the temperature factor sensitive load sequence and the non-temperature factor sensitive load sequence, respectively, taking into account the impact of temperature factors on the load, and thus the accuracy of the load scenario sequence obtained is higher, which improves the reliability of the long-term planning of the power system.

[0060] In an exemplary embodiment, the above step 201 divides the total load sequence of the target area throughout the year to obtain a temperature sensitive load sequence and a non-temperature sensitive load sequence, including steps 2011 to 2013, wherein:

[0061] Step 211, deleting the daily load sequence corresponding to holidays in the total load sequence to obtain the deleted total load sequence.

[0062] Among them, the obvious reduction in load during holidays such as the Spring Festival and National Day has little to do with meteorological factors. Therefore, this part of the data needs to be eliminated to improve the accuracy of the temperature change model.

[0063] For example, the daily load sequences corresponding to October 1 to October 7 in the total load sequence are deleted to obtain the deleted total load sequence.

[0064] Step 212, according to the geographical environment of the target area, the total load sequence after deletion is divided to obtain an initial temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence.

[0065] Among them, the geographical environment of the target area includes the climate, geography, and humanities of the target area, and the climate, geography, and humanities of different regions vary greatly. Therefore, it is necessary to divide the total load sequence after deletion according to the actual meteorological season, rather than dividing the total load sequence according to the calendar season. For example, the Northeast region has generated a large amount of electric heating demand in November, so the daily load sequence corresponding to November in the Northeast region is more suitable to be classified as winter rather than autumn, that is, it should be classified as the initial temperature factor sensitive load sequence.

[0066] Optionally, select multiple different characteristic indicators to perform characteristic analysis on the load sequences of each day in the total load sequence after deletion, and obtain characteristic analysis results of the load sequences of each day in the total load sequence after deletion; select multiple different meteorological factors to perform correlation analysis on the load sequences of each day in the total load sequence after deletion, and obtain correlation analysis results of the load sequences of each day in the total load sequence after deletion and different meteorological factors. According to the characteristic analysis results of the load sequences of each day in the total load sequence after deletion and the correlation analysis results of the load sequences of each day with different meteorological factors, the load sequences of each day in the total load sequence after deletion are divided to obtain the initial temperature factor sensitive load sequence and the non-temperature factor sensitive load sequence.

[0067] Step 213: based on the non-air temperature factor sensitive load sequence, the non-air temperature factor sensitive load contained in the initial air temperature factor sensitive load sequence is deleted to obtain the air temperature factor sensitive load sequence.

[0068] Among them, the initial temperature factor sensitive load sequence includes non-temperature factor sensitive load (also called basic load), temperature factor sensitive load and random load; among them, the proportion of random load is small and can be ignored.

[0069] In a possible implementation, multiple second-day load sequences in the non-temperature sensitive load sequence are divided in units of weeks to obtain multiple intra-weekly daily load sequences (daily load sequences from Monday to Friday) and multiple weekend daily load sequences (daily load sequences on Saturday and Sunday) corresponding to each week; a second preset number (hereinafter represented by m) of first weeks are extracted from the multiple weeks, and a third preset number (hereinafter represented by n) of second weeks are extracted from the multiple weeks. Among them, m is greater than or equal to 2, n is greater than or equal to 2, and m can be greater than n. For example, m can be 5, and n can be 3.

[0070] For each first week extracted, the smallest intra-week daily load sequence is selected from multiple intra-week daily load sequences of the first week, thereby obtaining m smallest intra-week daily load sequences, and the m smallest intra-week daily load sequences are averaged to obtain an intra-week average daily load sequence.

[0071] For each second week selected, the smallest weekend daily load sequence is selected from multiple weekend daily load sequences of the second week, thereby obtaining n smallest weekend daily load sequences, and the n smallest weekend daily load sequences are averaged to obtain a weekend average daily load sequence.

[0072] The average daily load series during the week and the average daily load series on the weekend are spliced ​​to obtain the basic load series.

[0073] In another possible implementation, multiple second-day load sequences in the non-temperature sensitive load sequence are divided to obtain multiple weekly daily load sequences (daily load sequences from Monday to Friday) and multiple weekend daily load sequences (daily load sequences on Saturday and Sunday). m minimum weekly daily load sequences are directly selected from multiple weekly daily load sequences, and the m minimum weekly daily load sequences are averaged to obtain the average weekly daily load sequence. n minimum weekend daily load sequences are selected from multiple weekend daily load sequences, and the n minimum weekend daily load sequences are averaged to obtain the average weekend daily load sequence.

[0074] The above process of obtaining the basic load sequence is expressed by formula (1) as follows:

[0075] (1)

[0076] in, represents the base load sequence, represents the daily load sequence from Monday to Friday, Represents the daily load sequence on Saturday and Sunday. Based on the above two implementation methods, It can represent the smallest intra-week daily load sequence corresponding to the ith week in the m first weeks, or it can represent the ith intra-week daily load sequence in the m smallest intra-week daily load sequences; It can represent the smallest weekend load sequence corresponding to the jth week among n second weeks; it can also represent the jth weekend load sequence among n smallest weekend load sequences.

[0077] Taking weeks as units, multiple daily load sequences in the initial temperature-sensitive load sequence are divided to obtain multiple daily load sequences corresponding to each week (daily load sequences from Monday to Sunday); for multiple daily load sequences corresponding to each week, the multiple daily load sequences corresponding to the targeted week are subtracted from the basic load sequence to obtain multiple first-day load sequences corresponding to the targeted week (daily load sequences from Monday to Sunday with non-temperature-sensitive loads deleted), thereby obtaining the temperature-sensitive load sequence. It can be understood that the difference between multiple daily load sequences and the basic load sequence is to subtract the daily load sequence belonging to the week from the average daily load sequence within the week in the basic load sequence, and to subtract the daily load sequence belonging to the weekend from the average daily load sequence on the weekend in the basic load sequence.

[0078] In one of the embodiments, before obtaining the temperature change model based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence, the method also includes: for each first-day load sequence in the temperature factor sensitive load sequence, based on a pre-constructed temperature cumulative effect optimization model, correcting the daily temperature corresponding to the first-day load sequence to obtain the daily target temperature corresponding to the first-day load sequence.

[0079] The daily temperature corresponding to the first day's load sequence is the daily average temperature corresponding to the first day's load sequence. The daily temperature corresponding to the first day's load sequence is obtained in advance. In a possible implementation, the terminal receives the total load sequence input by the user and the daily temperature corresponding to each daily load sequence in the total load sequence, so that the daily temperature corresponding to the first day's load sequence can be obtained.

[0080] Considering the cumulative effect of temperature, the daily temperature of the target day i (that is, the first day of the load sequence) is corrected, that is:

[0081] (2)

[0082] in, is the daily target temperature of target day i after considering the cumulative effect of temperature; and are the average temperature of the j days before the target day i and its corresponding cumulative effect coefficient, and p is the maximum cumulative number of days. The maximum cumulative number of days and the cumulative effect coefficient are iteratively solved. Under the selected maximum cumulative number of days p, the temperature cumulative effect optimization model is constructed to solve the corresponding cumulative effect coefficient sequence. The constructed temperature cumulative effect optimization model is shown in formula (3) and formula (4).

[0083] (3)

[0084] (4)

[0085] Among them, formula (3) is the objective function, and formula (4) is the constraint condition; is the maximum daily temperature change load on the i-th day, that is, the maximum load in the first-day load sequence corresponding to the i-th day; is the standard deviation, is the covariance. Select the correlation coefficient The p value corresponding to the maximum value of is taken as the optimal solution for the maximum cumulative number of days, and the corresponding As the optimal solution of the cumulative effect coefficient sequence.

[0086] In an exemplary embodiment, a multivariate linear regression model is used to sequentially fit the daily target temperatures corresponding to multiple first-day load sequences of each historical year and the daily maximum temperature change loads corresponding to multiple first-day load sequences to obtain a temperature change model for each year. The temperature change model for each year is shown in formula (5).

[0087] (5)

[0088] Where y is the year, is the maximum daily temperature change load on the i-th day of the y-th year, is the daily target temperature on the i-th day of the y-th year after considering the cumulative effect of temperature, To consider the characteristic index of the weekend factor, the weekdays are recorded as 0 and the weekends are recorded as 1; , and is the regression coefficient calculated based on the least squares method; is the residual.

[0089] Characterizes the sensitivity of the maximum daily temperature load in year y to the temperature change, which can be called the temperature sensitivity coefficient; It is related to the electrification level of the target area, and the relationship is positive. Therefore, the temperature sensitivity coefficient of the target area in each year can be obtained by fitting the daily target temperature corresponding to multiple first-day load sequences and the daily maximum temperature change load corresponding to multiple first-day load sequences. Then, the positive correlation between the temperature sensitivity coefficient and the electrification level can be modeled based on the temperature sensitivity coefficient of the target area in each year and the electrification level of the target area in each year, and the temperature sensitivity coefficient prediction model can be obtained, that is, formula (6).

[0090] (6)

[0091] in, is the electrification level of the target area in year y, which can be measured by the proportion of electricity in final energy consumption.

[0092] Therefore, generating the first load scenario sequence according to the temperature change model in step 202 may include the following steps 2021 to 2024, wherein:

[0093] Step 2021, correcting the temperature sensitivity coefficient in the temperature change model according to the electrification level in the year to be predicted, to obtain a corrected temperature change model.

[0094] Optionally, the electrification level of the year to be predicted is input into the temperature sensitivity coefficient prediction model, i.e., formula (6), to obtain the temperature sensitivity coefficient of the year to be predicted. The temperature sensitivity coefficient in the above formula (5) is replaced based on the temperature sensitivity coefficient of the year to be predicted, and the temperature change model corresponding to the year to be predicted can be obtained, that is, the corrected temperature change model.

[0095] In addition, in formula (5) Characterizing the volatility and uncertainty of load, nonparametric kernel density estimation can be used to estimate the residual The probability density fitting is performed on the distribution of the predicted years, so the residuals in the temperature change model corresponding to the predicted years can be further corrected based on the probability density fitting results.

[0096] Step 2022, input the temperature of the day to be predicted into the corrected temperature change model to obtain the daily maximum temperature change load of the day to be predicted.

[0097] The daily temperatures corresponding to the daily load sequences in the total load sequence can be sequenced to obtain a daily temperature sequence, and the temperature of the day to be predicted can be obtained based on the daily temperature sequence. For example, the temperature on November 1, 2022, the year to be predicted, can be obtained based on the temperature corresponding to November 1 in the daily temperature sequence of 2021.

[0098] Step 2023: select a first preset number (hereinafter denoted by q) of first-day load sequences from the temperature factor sensitive load sequence according to the daily maximum temperature variation load of the day to be predicted. q is greater than or equal to 2, for example, it can be 3.

[0099] Step 2024: Obtain a first load scenario sequence based on a first preset number of first-day load sequences selected.

[0100] Optionally, the temperature of the day to be predicted is input into the revised temperature change model to obtain the daily maximum temperature change load of the day to be predicted, and then q first-day load sequences with the smallest difference from the daily maximum temperature change load of the day to be predicted are selected from multiple first-day load sequences.

[0101] The q selected first-day load sequences are averaged every 24 hours to obtain a reference load sequence; then, according to the peak value of the reference load sequence and the multiple of the daily maximum temperature change load of the predicted day, the reference load sequence is scaled proportionally to obtain the first load scenario sequence.

[0102] The above process of obtaining the first load scenario sequence is expressed as follows using formula (7) and formula (8):

[0103] (7)

[0104] (8)

[0105] in, is the load at time t on q first-day load sequences selected from multiple first-day load sequences; is the load at time t in the reference load sequence averaged over 24 hours; is the daily maximum temperature change load on the predicted day i; is the daily load sequence of day i to be predicted, which is also the first load scenario sequence.

[0106] The above describes in detail the process of generating the first load scenario sequence. The following describes in detail the process of generating the second load scenario sequence.

[0107] In an exemplary embodiment, the step 203 of generating the second load scenario sequence according to the joint probability distribution model includes steps 2031 to 2033, wherein:

[0108] Step 2031, sampling from the joint probability distribution model to obtain multiple different characteristic indicators of the day to be predicted.

[0109] Optionally, the joint probability distribution model is randomly sampled to obtain a plurality of different characteristic indicators of the day to be predicted, such as at least two of the daily load mean, peak value, peak-to-valley difference, peak load rate, and valley load rate. Preferably, the plurality of different characteristic indicators are the daily load mean, peak value, peak-to-valley difference, peak load rate, and valley load rate.

[0110] Step 2032: According to a plurality of different characteristic indicators of the day to be predicted, a reference load sequence is obtained by screening from a plurality of second-day load sequences of the non-temperature factor sensitive load sequence.

[0111] Optionally, multiple different characteristic indicators of each second-day load sequence are obtained, and for each second-day load sequence, similarity is calculated between the multiple different characteristic indicators of the second-day load sequence and the multiple different characteristic indicators of the day to be predicted to obtain the similarity of the second-day load sequence; and the second-day load sequence with the greatest similarity is selected from the multiple second-day load sequences as the benchmark load sequence.

[0112] The similarity calculation may be performed by using a cosine similarity method or a Euclidean distance method.

[0113] Step 2033: obtaining a second load scenario sequence based on the pre-built load sequence optimization model and the benchmark load sequence.

[0114] The pre-built load sequence optimization model is as shown in formula (9) and formula (10), namely:

[0115] (9)

[0116] (10)

[0117] Among them, formula (9) is the objective function, and formula (10) is the constraint condition. The objective function is to minimize the sum of squares of the errors between the generated daily load sequence (that is, the second load scenario sequence) and the benchmark load sequence, so as to ensure that the curve trend between the generated daily load sequence and the benchmark load sequence is closest. is the generated daily load series; is the benchmark load sequence obtained by screening. The basic idea of ​​the constraint condition is that the characteristic index of the generated daily load sequence is consistent with the characteristic index obtained by sampling. to They represent the five characteristic indicators obtained by sampling, namely, the daily load mean, peak value, peak-to-valley difference, peak load rate and valley load rate; and They represent the moments when the maximum load and the minimum load occur in the benchmark load sequence respectively; and represent the load peak period and the number of load peak periods respectively, and They represent the load valley period and the number of load valley periods respectively.

[0118] In one embodiment, Figure 3 As shown, the present application also provides a schematic diagram of the principle of generating a load scenario sequence. The terminal obtains the total load sequence of the target area (that is, the hourly load data) and the daily temperature corresponding to each daily load sequence in the total load sequence (that is, the daily temperature data). The terminal generates a sequence of the daily temperatures corresponding to each daily load sequence in the total load sequence to obtain a daily temperature sequence, and the terminal performs feature analysis and correlation analysis on the total load sequence, and divides the total load sequence into two categories, one is an initial temperature factor sensitive load sequence that is sensitive to temperature, and the other is a non-temperature factor sensitive load sequence that is insensitive to temperature.

[0119] For the initial temperature sensitive load sequence, delete the non-temperature sensitive loads in the initial temperature sensitive load sequence to obtain the temperature sensitive load sequence. Consider the temperature accumulation effect to correct the daily temperature corresponding to each first-day load sequence in the temperature sensitive load sequence to obtain the daily target temperature corresponding to each first-day load sequence. Based on the daily target temperature corresponding to each first-day load sequence in the temperature sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence, a temperature change model is obtained. The temperature of the day to be predicted is obtained based on the daily temperature sequence, and the first load scenario sequence is generated based on the temperature and temperature change model of the day to be predicted.

[0120] For the non-temperature sensitive load sequence, multiple different characteristic indicators of each second-day load sequence in the non-temperature sensitive load sequence are counted. For each characteristic indicator, the non-parametric kernel density estimation method is used to fit the characteristic indicators targeted by multiple second-day load sequences to obtain the marginal probability distribution of the targeted characteristic indicators. A joint probability distribution model is constructed based on the marginal probability distribution of multiple characteristic indicators. Multiple different characteristic indicators of the day to be predicted are sampled from the joint probability distribution model. According to the multiple different characteristic indicators of the day to be predicted, a benchmark load sequence is screened from multiple second-day load sequences of the non-temperature sensitive load sequence. A load sequence optimization model is constructed, and the second load scenario sequence is obtained based on the constructed load sequence optimization model and the benchmark load sequence.

[0121] Through the above method, multiple first load scenario sequences and multiple second load scenarios for the whole year of the year to be predicted can be obtained, that is, the load scenario sequence for the whole year of the year to be predicted can be obtained.

[0122] The specific implementation of this embodiment has been described in detail above and will not be repeated here.

[0123] In summary, this application generates scenario sequences for load sequences for medium- and long-term power system planning while taking into account the influence of electrification levels and temperature factors, thereby achieving effective modeling of future power load development trends and improving the reliability of medium- and long-term planning and operation analysis results of the power system.

[0124] The following analysis is based on three provincial power grids, as follows:

[0125] (1) Load characteristics analysis and total load sequence division.

[0126] The correlation coefficients between the daily maximum load and meteorological factors in different seasons in Province 1 are shown in Table 1. It can be seen from Table 1 that the daily average temperature has the largest positive / negative correlation coefficient and can be used as the primary meteorological factor affecting the load in summer and winter.

[0127] Table 1

[0128]

[0129] The comparison of the correlation coefficients between the daily maximum temperature change load and the daily average temperature before and after temperature correction in each province from 2015 to 2019 in summer is shown in Table 2. It can be seen that considering the cumulative effect of temperature can effectively improve the correlation between load and temperature indicators.

[0130] Table 2

[0131]

[0132] Summer temperature sensitivity coefficient of each province year by year With the level of electrification The changing trend of Figure 4 As shown in Figure 2, it can be seen that as the electrification level of the region increases, the sensitivity of the load to temperature continues to increase, and the two are significantly positively correlated. After reaching a certain level, the temperature sensitivity coefficient The growth rate has slowed down. Figure 4 The horizontal axis is the electrification level, and the vertical axis is the temperature sensitivity coefficient.

[0133] (2) Generation of load scenarios without considering the impact of electrification level and temperature factors.

[0134] Without considering the influence of electrification level and temperature factors, the distribution of characteristic indicators of historical load series and load scenario series generated by this application in different seasons is compared. Figure 5 The probability density PDF curves of the historical load series of each season and the load scenario series generated by this application are shown in Figure 6 As shown, it can be seen that this application can well capture the distribution characteristics of historical load series by modeling and analyzing the loads in different seasons. Figure 5 The horizontal axis represents different seasons, and the vertical axis represents probability; Figure 6 The horizontal axis is the load, and the vertical axis is the probability density PDF. The pu in the figure is a dimensionless unit used to represent relative values ​​in the power system, that is, the per unit value, the same below.

[0135] (3) Generation of load scenarios considering the impact of electrification level and temperature factors.

[0136] The temperature sensitivity coefficient and temperature series are adjusted to characterize the future load development trend. Figure 7 The distribution of summer load scenario sequences under three different development scenarios was compared. Among them, the meaning of the frequent extreme high temperature scenario (scenario 1) is: the regional electrification level remains unchanged, and there is a new high temperature weather with a weekly average temperature exceeding 30°C in summer; the meaning of the electrification level improvement scenario (scenario 2) is: the regional electrification level is improved, and the summer cooling load accounts for 80% of the maximum power load, using the historical temperature sequence; the meaning of the high temperature superimposed on the electrification improvement scenario (scenario 3) is: the regional electrification level is improved and the continuous high temperature weather is superimposed, and the temperature sensitivity coefficient of the load and the parameter settings of the local daily average temperature sequence are consistent with scenario 2 and scenario 1, respectively. It can be seen that due to the influence of continuous high temperature weather, the daily load peak increases, and the corresponding probability density curve shows a rightward shift trend; due to the improvement of the regional electrification level, the proportion of temperature-variable load in the total load increases, and the load is more sensitive to temperature changes. Therefore, the fluctuation range of the daily load peak is wider and the load uncertainty is stronger; the continuous high temperature superimposed on the electrification improvement scenario will cause the load peak to further increase. Among them, Figure 7 The horizontal axis is the daily load peak, and the vertical axis is the probability density PDF.

[0137] Figure 8 The summer daily maximum load sequence and hourly load average under a week of continuous high temperature weather corresponding to the historical scenario and scenario 3 are given. Affected by the continuous high temperature weather, the load side in scenario 3 also produced an obvious load peak in early July, and the load peak effect caused by the continuous high temperature weather in summer was more prominent due to the increase in the proportion of temperature-dependent loads. Figure 8 In (a), the horizontal axis is the date, and the vertical axis is the daily maximum load; Figure 8 In Figure (b), the horizontal axis is hours and the vertical axis is load.

[0138] The impact of load scenario sequences on the operation mode of thermal power units in power systems is as follows: Fig. 9 As shown, the annual load sequence of scenario 1 is obtained by proportionally amplifying the historical load sequence based on the annual electricity forecast results; the annual load sequence of scenario 2 adopts the load scenario sequence generation method of this application, taking into account the impact of the improvement of regional electrification level and the new continuous high temperature weather in summer. It can be seen that due to the impact of the improvement of electrification level and continuous high temperature weather, the online capacity of thermal power units needs to be maintained at a higher level to ensure a reliable supply of electricity in summer. In addition, in load scenario 2, the sum of the summer utilization hours and start-stop capacity of coal-fired units and gas-fired units has increased, indicating that the increase in summer peak load requires thermal power units to continue to assume the task of ensuring power supply. At the same time, gas-fired units need to be used as flexibility resources to provide the system with stronger flexible adjustment capabilities to adapt to a wider range of load fluctuations in the future. The operating mode of the system in scenario 2 is more diverse. Fig. 9 The horizontal axis is the online capacity of thermal power units, in GW, and the vertical axis is the cumulative probability CDF.

[0139] It can be seen from the above data that when the interannual changes of load characteristics and temperature factors are not considered, this application can effectively simulate the probability distribution characteristics of historical loads and the distribution characteristics of different seasons; when considering the development trend of increasing continuous high temperature weather in summer and the increased sensitivity of load to temperature changes, this application can reasonably extrapolate and correct the model parameters based on the load and temperature data of many years in history, and fully characterize the changing trends of load characteristics and power system operation mode under given future development scenarios. This method will have a significant effect on the generation of load scenario sequences suitable for medium- and long-term power system planning and operation analysis. It has clear calculation ideas and good versatility, and is suitable for promotion and use.

[0140] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0141] Based on the same inventive concept, the embodiment of the present application also provides a load scenario sequence generation device for implementing the load scenario sequence generation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more load scenario sequence generation device embodiments provided below can refer to the limitations of the load scenario sequence generation method above, and will not be repeated here.

[0142] In an exemplary embodiment, Fig.10 As shown, a load scenario sequence generation device is provided, comprising: a division module 1001, a first generation module 1002 and a second generation module 1003, wherein:

[0143] The division module 1001 is used to divide the total load sequence of the target area throughout the year to obtain a temperature-sensitive load sequence and a non-temperature-sensitive load sequence.

[0144] The first generating module 1002 is used to obtain a temperature variation model based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature variation load corresponding to each first-day load sequence; and generate a first load scenario sequence according to the temperature variation model.

[0145] The second generation module 1003 is used to count multiple different characteristic indicators of each second-day load sequence in the non-temperature factor sensitive load sequence to obtain a joint probability distribution model; and generate a second load scenario sequence according to the joint probability distribution model.

[0146] In one embodiment, the device also includes a correction module for correcting the daily temperature corresponding to the first-day load sequence in the temperature-sensitive load sequence based on a pre-built temperature cumulative effect optimization model to obtain the daily target temperature corresponding to the first-day load sequence.

[0147] In one embodiment, the temperature change model includes a temperature sensitivity coefficient. The first generation module 1001 is specifically used to correct the temperature sensitivity coefficient in the temperature change model according to the electrification level of the year to be predicted to obtain a corrected temperature change model; the temperature of the day to be predicted is input into the corrected temperature change model to obtain the daily maximum temperature change load of the day to be predicted; according to the daily maximum temperature change load of the day to be predicted, a first preset number of first-day load sequences are selected from the temperature factor sensitive load sequence; and a first load scenario sequence is obtained based on the selected first preset number of first-day load sequences.

[0148] In one embodiment, the second generation module 1003 is specifically used to sample multiple different characteristic indicators of the day to be predicted from the joint probability distribution model; based on the multiple different characteristic indicators of the day to be predicted, select a benchmark load sequence from multiple second-day load sequences of the non-temperature factor sensitive load sequence; based on the pre-constructed load sequence optimization model and the benchmark load sequence, obtain a second load scenario sequence.

[0149] In one embodiment, the second generation module 1003 is specifically used to obtain multiple different characteristic indicators of each second-day load sequence, and for each second-day load sequence, perform similarity calculations on multiple different characteristic indicators of the second-day load sequence and multiple different characteristic indicators of the day to be predicted to obtain the similarity of the second-day load sequence; and select the second-day load sequence with the greatest similarity from multiple second-day load sequences as the benchmark load sequence.

[0150] In one embodiment, the division module 1001 is specifically used to delete the daily load sequence corresponding to holidays in the total load sequence to obtain the total load sequence after deletion; divide the total load sequence after deletion according to the geographical environment of the target area to obtain an initial temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence; based on the non-temperature factor sensitive load sequence, delete the non-temperature factor sensitive load contained in the initial temperature factor sensitive load sequence to obtain a temperature factor sensitive load sequence.

[0151] Each module in the above-mentioned load scenario sequence generation device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0152] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Fig.11As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a load scenario sequence generation method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0153] Those skilled in the art will understand that Fig.11 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0154] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps of any one of the method embodiments described above are implemented.

[0155] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the above method embodiments are implemented.

[0156] In one embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements the steps of any one of the method embodiments described above.

[0157] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0158] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0159] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for generating a load scenario sequence, characterized in that: The method comprises: The total load sequence of the target area throughout the year is divided to obtain the temperature-sensitive load sequence and the non-temperature-sensitive load sequence; Based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence, a temperature change model is obtained; and a first load scenario sequence is generated according to the temperature change model; A plurality of different characteristic indicators of each second-day load sequence in the non-temperature factor sensitive load sequence are counted to obtain a joint probability distribution model; and a second load scenario sequence is generated according to the joint probability distribution model.

2. The method according to claim 1, characterized in that Before obtaining the temperature change model based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence, the method further includes: For each first-day load sequence in the temperature-sensitive load sequence, the daily temperature corresponding to the first-day load sequence is corrected based on a pre-constructed temperature cumulative effect optimization model to obtain a daily target temperature corresponding to the first-day load sequence.

3. The method according to claim 1, characterized in that The temperature variation model includes a temperature sensitivity coefficient, and generating a first load scenario sequence according to the temperature variation model includes: Correcting the temperature sensitivity coefficient in the temperature change model according to the electrification level in the year to be predicted, to obtain a corrected temperature change model; Inputting the temperature of the day to be predicted into the modified temperature change model to obtain the daily maximum temperature change load of the day to be predicted; Selecting a first preset number of first-day load sequences from the temperature factor sensitive load sequence according to the daily maximum temperature change load of the day to be predicted; The first load scenario sequence is obtained according to the selected first preset number of first-day load sequences.

4. The method according to claim 1, characterized in that: The generating a second load scenario sequence according to the joint probability distribution model comprises: Sampling from the joint probability distribution model to obtain a plurality of different characteristic indicators of the day to be predicted; According to a plurality of different characteristic indicators of the day to be predicted, a reference load sequence is obtained by screening from a plurality of second-day load sequences of the non-temperature factor sensitive load sequence; Based on the pre-constructed load sequence optimization model and the benchmark load sequence, the second load scenario sequence is obtained.

5. The method according to claim 4, characterized in that The method of selecting a reference load sequence from a plurality of second-day load sequences of the non-temperature factor sensitive load sequence according to a plurality of different characteristic indicators of the day to be predicted includes: Obtaining a plurality of different characteristic indicators of each second-day load sequence, For each of the second-day load sequences, similarity calculation is performed on a plurality of different characteristic indicators of the second-day load sequence and a plurality of different characteristic indicators of the day to be predicted, so as to obtain the similarity of the second-day load sequence; The second-day load sequence with the greatest similarity is selected from the plurality of the second-day load sequences as the reference load sequence.

6. The method according to any one of claims 1 to 5, characterized in that: The total load sequence of the target area throughout the year is divided to obtain a temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence, including: Deleting the daily load sequence corresponding to holidays in the total load sequence to obtain the deleted total load sequence; According to the geographical environment of the target area, the deleted total load sequence is divided to obtain an initial temperature factor sensitive load sequence and a non-temperature factor sensitive load sequence; Based on the non-air temperature factor sensitive load sequence, the non-air temperature factor sensitive loads contained in the initial air temperature factor sensitive load sequence are deleted to obtain the air temperature factor sensitive load sequence.

7. A load scenario sequence generation device, characterized in that: The device comprises: A division module is used to divide the total load sequence of the target area throughout the year to obtain a temperature-sensitive load sequence and a non-temperature-sensitive load sequence; A first generating module is used to obtain a temperature change model based on the daily target temperature corresponding to each first-day load sequence in the temperature factor sensitive load sequence and the daily maximum temperature change load corresponding to each first-day load sequence; and generate a first load scenario sequence according to the temperature change model; The second generation module is used to count multiple different characteristic indicators of each second-day load sequence in the non-temperature factor sensitive load sequence to obtain a joint probability distribution model; and generate a second load scenario sequence according to the joint probability distribution model.

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

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

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images