Distributed photovoltaic aggregator day-ahead declaration power determination method and device, computer equipment and storage medium

Abstract

The invention provides a distributed photovoltaic aggregator day-ahead declaration power determination method and device, computer equipment and a storage medium. According to the method, a similar historical day with the highest reference value is selected from a plurality of historical days in a preset time span according to the similarity of load information, the positive price difference probability and the negative price difference probability of the similar historical day are taken as a reference, and on the basis of predicted power of each target time-sharing, reduction or increase is carried out, so that the prediction efficiency is improved. And obtaining the day-ahead declaration power of each target time division. According to the method, the day-ahead declaration power can be automatically determined after the day-ahead information is issued in the electricity transaction center, a scientific basis is provided for market participants to make decisions, the declaration risk is reduced, the possibility of income is improved, the working efficiency of distributed photovoltaic aggregators is also improved, and the workload brought by manual accounting is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of power market, and in particular to a method, device, computer equipment and storage medium for determining the power reported by a distributed photovoltaic aggregator on the previous day. Background Art

[0002] With the rapid development of new energy technologies, the problem of power consumption has become increasingly prominent. The participation of new energy in the electricity spot market will be one of the main ways to consume in the future. Distributed photovoltaic aggregators are one of the ways to achieve photovoltaic participation in the electricity market. Distributed photovoltaic aggregators currently declare all the predicted power in the day-ahead market and provide the declared power to the power market in the real-time market. However, this method will cause large losses to distributed photovoltaic aggregators due to power forecast deviations and electricity price fluctuations. Therefore, how to reasonably determine the day-ahead power declared by distributed photovoltaic aggregators is an issue that needs to be urgently addressed in the development of new energy marketization. Summary of the invention

[0003] The purpose of this application is to solve at least one of the above-mentioned technical defects, and in particular to provide a solution for determining the day-ahead power reported by distributed photovoltaic aggregators based on high efficiency.

[0004] In the first aspect, the present application provides a method for determining the power reported by a distributed photovoltaic aggregator on the previous day, including:

[0005] Obtain historical electricity price information and historical load information corresponding to each time period for multiple historical days within a preset time span; historical electricity price information includes historical day-ahead electricity price and historical real-time electricity price;

[0006] Determine each time-sharing as the target time-sharing, and determine the predicted power at each target time-sharing on the target trading day;

[0007] Match the load information published on the target trading day before the target time-sharing date with the historical load information of each historical day at the target time-sharing date, and select the historical days corresponding to the most similar preset number of historical load information as the similar historical days corresponding to the target time-sharing date;

[0008] According to the historical electricity price information corresponding to the target time-sharing on each similar historical day corresponding to the target time-sharing, the positive price difference probability and the negative price difference probability of the target time-sharing are determined;

[0009] If the positive price difference probability is greater than the first threshold, the predicted power corresponding to the target time is reduced to obtain the day-ahead declared power corresponding to the target time;

[0010] If the probability of negative price difference is greater than the first threshold, it is increased based on the predicted power corresponding to the target time period to obtain the day-ahead declared power corresponding to the target time period.

[0011] In one embodiment, the historical load information includes a variety of other historical loads that affect the market-based load, and the day-ahead load information includes a variety of other day-ahead loads that affect the market-based load;

[0012] The load information published on the target trading day before the target time-sharing date is matched with the historical load information of each historical day at the target time-sharing date, and the historical days corresponding to the most similar preset number of historical load information are selected as similar historical days, including:

[0013] For any historical day, determine the first similarity between the historical market-based load at the target time and the market-based load on the day before, and then determine the second similarities between other historical loads and other day-ahead loads of the same type, and obtain the third similarity corresponding to the historical day based on the first similarity, each second similarity and its corresponding first weight; the first weight corresponding to the first similarity is the largest, and the first weight corresponding to the second similarity obtained based on the type with the greater impact on the market-based load is greater;

[0014] A preset number of historical days are selected as similar historical days in descending order of the third similarity.

[0015] In one embodiment, if the positive price difference probability is greater than the first threshold, the predicted power corresponding to the target time is reduced to obtain the day-ahead declared power corresponding to the target time, including:

[0016] When the positive price difference probability is greater than the first threshold, determining a first target interval to which the positive price difference probability belongs among a plurality of consecutive first intervals;

[0017] Determine the target reduction according to the first mapping relationship and the first target interval; the first mapping relationship reflects the one-to-one correspondence between each first interval and the optional reduction, and the larger the probability value contained in the first interval, the larger the corresponding optional reduction;

[0018] According to the target reduction rate, the predicted power is reduced to obtain the day-ahead power corresponding to the target time period.

[0019] In one embodiment, if the probability of negative price difference is greater than the first threshold, the predicted power corresponding to the target time is increased to obtain the day-ahead declared power corresponding to the target time, including:

[0020] When the negative price difference probability is greater than the first threshold, determining a second target interval to which the negative price difference probability belongs among a plurality of consecutive second intervals;

[0021] Determine the target increase according to the second mapping relationship and the second target interval; the second mapping relationship reflects the one-to-one correspondence between each second interval and the optional increase, and the larger the probability value contained in the second interval, the larger the corresponding optional increase;

[0022] According to the target increase, the predicted power is increased to obtain the day-ahead power corresponding to the target time period.

[0023] In one embodiment, determining the predicted power at each target time on the target trading day includes:

[0024] Power prediction is performed according to a plurality of meteorological forecast data of a target time segment and the corresponding second weights to obtain the predicted power of the target time segment.

[0025] In one embodiment, power prediction is performed according to multiple meteorological forecast data of the target time and their corresponding second weights to obtain the predicted power of the target time, including:

[0026] Determine the forecast accuracy of various weather forecast data within a preset time span;

[0027] For any kind of weather forecast data, an adjustment coefficient is determined according to the forecast accuracy and the third mapping relationship, and the target adjustment coefficient is multiplied by the initial weight corresponding to the weather forecast data to obtain a second weight corresponding to the weather forecast data; the third mapping relationship reflects the mapping relationship between the forecast accuracy and the adjustment coefficient, the adjustment coefficient is less than or equal to 1, and the higher the forecast accuracy, the closer the corresponding adjustment coefficient is to 1;

[0028] Power prediction is performed according to a plurality of meteorological forecast data of a target time segment and the corresponding second weights to obtain the predicted power of the target time segment.

[0029] In one embodiment, according to the historical electricity price information corresponding to the target time-sharing on each similar historical day corresponding to the target time-sharing, the positive price difference probability and the negative price difference probability of the target time-sharing are determined, including:

[0030] If the historical real-time electricity price at the target time-sharing time on a similar historical day is higher than the historical day-ahead electricity price, the similar historical day is classified as a positive price difference, otherwise, it is classified as a negative price difference;

[0031] The probability of positive price difference is obtained based on the ratio of the number of similar historical days with positive price difference to the preset number;

[0032] The probability of negative price spread is obtained based on the ratio of the number of similar historical days with negative price spread to the preset number.

[0033] In the second aspect, the present application provides a device for determining the power reported by a distributed photovoltaic aggregator on the previous day, comprising:

[0034] A data acquisition module is used to obtain historical electricity price information and historical load information corresponding to each time period for multiple historical days within a preset time span; the historical electricity price information includes historical day-ahead electricity prices and historical real-time electricity prices;

[0035] A power prediction module is used to determine each time-sharing as a target time-sharing and determine the predicted power at each target time-sharing on the target trading day;

[0036] A matching module is used to match the load information published on the target trading day before the target time-sharing date with the historical load information of each historical day at the target time-sharing date, and select the historical days corresponding to the most similar preset number of historical load information as the similar historical days corresponding to the target time-sharing date;

[0037] A probability determination module is used to determine the positive price difference probability and negative price difference probability of the target time-sharing according to the historical electricity price information corresponding to the target time-sharing on each similar historical day;

[0038] The first adjustment module is used for reducing the predicted power corresponding to the target time-sharing period to obtain the day-ahead declared power corresponding to the target time-sharing period when the positive price difference probability is greater than the first threshold value;

[0039] The second adjustment module is used to increase the predicted power corresponding to the target time-sharing when the probability of negative price difference is greater than the first threshold, so as to obtain the day-ahead declared power corresponding to the target time-sharing.

[0040] In a third aspect, the present application provides a computer device comprising one or more processors and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the steps of the method for determining the day-ahead power reported by a distributed photovoltaic aggregator in any of the above-mentioned embodiments are executed.

[0041] In a fourth aspect, the present application provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the method for determining the day-ahead power reported by a distributed photovoltaic aggregator in any of the above-mentioned embodiments.

[0042] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0043] This method selects the most valuable similar historical days from multiple historical days within a preset time span based on the similarity of load information, and takes the positive price difference probability and negative price difference probability of similar historical days as a reference, and reduces or increases the predicted power of each target time period to obtain the day-ahead declared power of each target time period. This method can automatically determine the day-ahead declared power after the power trading center releases the day-ahead information, providing a scientific basis for market participants to make decisions, reducing declaration risks, and increasing the possibility of benefits. It also improves the work efficiency of distributed photovoltaic aggregators and reduces the workload caused by manual accounting. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art 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 drawings can be obtained based on these drawings without paying creative labor.

[0045] Figure 1 This is a flow chart of a method for determining the power reported by a distributed photovoltaic aggregator in one embodiment of the present application;

[0046] Figure 2 A schematic diagram of a process for determining similar historical days in one embodiment of the present application;

[0047] Figure 3 A schematic diagram of a process for adjusting the predicted power to obtain the day-ahead declared power when the positive price difference probability is greater than a first threshold in one embodiment of the present application;

[0048] Figure 4 A schematic diagram of a process for adjusting the predicted power to obtain the day-ahead declared power when the probability of a negative price difference is greater than a first threshold in one embodiment of the present application;

[0049] Figure 5 A schematic diagram of a flow chart for obtaining the predicted power of each target time-sharing in one embodiment of the present application;

[0050] Figure 6 An internal structure diagram of a computer device provided for one embodiment of the present application. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0052] This application provides a method for determining the power reported by a distributed photovoltaic aggregator on the previous day. Figure 1 , including steps S102 to S112.

[0053] S102, obtaining historical electricity price information and historical load information corresponding to each time period for multiple historical days within a preset time span. The historical electricity price information includes historical day-ahead electricity prices and historical real-time electricity prices.

[0054] It can be understood that the preset time span refers to a pre-selected period of time, and each day within the range is called a historical day. This embodiment will analyze based on some data of historical days. The preset time span can be from the date when the spot market starts to run to the date when the method in this embodiment is run.

[0055] The data of each historical day is specifically the historical electricity price information and historical load information of each time period of the historical day. The historical electricity price information includes the historical day-ahead electricity price and the historical real-time electricity price. The spot market includes the day-ahead market and the real-time market. The day-ahead market is a market for electricity trading based on the load forecast and power generation capacity assessment for the next day. The real-time market is the electricity trading conducted at the actual operation time of the power system. The historical day-ahead electricity price is the electricity price in the day-ahead market, and the historical real-time electricity price is the electricity price in the real-time market.

[0056] Historical load information refers to the time-sharing load information corresponding to historical days. The load information here can include the total load dispatched by the entire power system, that is, the centralized load, and multiple components of the centralized load, such as market-based load (i.e., load that enters the market-based electricity transaction), non-market-based load (i.e., load that does not obtain the qualification for the market-based electricity transaction), local load, etc.

[0057] In the electricity market, understanding historical data is crucial for forecasting and decision-making. By obtaining historical electricity price information and historical load information at various times of multiple historical days within a preset time span. The historical day-ahead electricity price can help understand the difference between the day-ahead electricity price and the real-time electricity price, so as to make scientific decisions. Since the electricity demand situation will determine the electricity price to a certain extent, the historical load information can reflect the electricity demand situation in different time periods, and the electricity prices of trading days with similar load conditions will also be similar. The historical load information can help select the most valuable reference parts from many historical days.

[0058] S104, respectively determining each time-sharing as a target time-sharing, and determining the predicted power at each target time-sharing on the target trading day.

[0059] It can be understood that the target trading day is the second day of the day-ahead power declaration, that is, the day when the real-time market is conducted. In this embodiment, the power size that needs to be declared is determined for each time-sharing, and each time-sharing will be selected as the target time-sharing to carry out subsequent steps separately. First, it is necessary to determine the predicted power of the distributed photovoltaics managed by the distributed photovoltaic aggregator at each target time-sharing. Predicting the power of distributed photovoltaics is currently a hot research direction, and mature algorithms can be selected for use. Generally speaking, distributed photovoltaic predictions need to consider a variety of meteorological forecast data corresponding to the target time-sharing, such as weather type, sunshine intensity, temperature, humidity, wind speed, etc., and assign different weights for power prediction. The traditional method is to declare the predicted power in full in the day-ahead market, and this application will make adjustments based on the predicted power to help distributed photovoltaic aggregators increase their profits.

[0060] S106, matching the load information published on the target trading day before the target time-sharing day with the historical load information of each historical day at the target time-sharing day, and selecting the historical days corresponding to the most similar preset number of historical load information as the similar historical days corresponding to the target time-sharing day.

[0061] It is understandable that the power trading center will disclose the day-ahead information in the day-ahead market, including the power trading center's forecast of the load situation on the next day, that is, the day-ahead load information. The day-ahead load information can include both forecasts for the unified load and forecasts for multiple components of the unified load. Since the electricity prices on trading days with similar load conditions will also be similar, for each target time-sharing, the day-ahead load information is matched with the historical load information of other historical days at the same time-sharing, and the similarity of the load situation on the target trading day and each historical day at the target time-sharing can be determined. The higher the similarity, the more reference value the electricity price situation on the target time-sharing day on the historical day has. Therefore, the most similar preset number (such as 30) can be selected as historical similar days. Subsequently, the overall electricity price situation on historical similar days will be used as a reference to adjust the power declaration for the target time-sharing.

[0062] S108, determining the positive price difference probability and the negative price difference probability of the target time-sharing according to the historical electricity price information corresponding to the target time-sharing on each similar historical day.

[0063] It can be understood that if the real-time electricity price is greater than the day-ahead electricity price, it is called a positive price difference, otherwise it is called a negative price difference. The overall electricity price situation mentioned above is mainly to count the probability of positive price differences and the probability of negative price differences in all similar historical days. For example, if there are six similar historical days with positive price differences in the target time-sharing period out of ten similar historical days, then the probability of positive price differences is 60%. By determining the probability of positive price differences and negative price differences, the possibility of positive and negative price differences occurring in the target time-sharing period can be understood. Since the real-time electricity price and the day-ahead electricity price are rarely exactly the same, the general sum of the probability of positive price differences and the probability of negative price differences is 1. If the probability of positive price differences is greater, go to step S110, and if the probability of negative price differences is greater, go to step S112. Of course, if the probabilities of positive and negative price differences are exactly equal, they can be declared directly according to the predicted power.

[0064] S110, if the positive price difference probability is greater than the first threshold, then the predicted power corresponding to the target time is reduced to obtain the day-ahead declared power corresponding to the target time.

[0065] It can be understood that the first threshold is a pre-set judgment standard value. Generally, the first threshold can be selected as 50%. If the probability of a positive price difference is greater than the first threshold, it means that the probability of a positive price difference is greater than that of a negative price difference. It also means that the possibility of a real-time electricity price higher than the day-ahead electricity price at the target time is greater. When the probability of a positive price difference is greater than the first threshold, declaring all predicted powers will result in all predicted powers being sold at the day-ahead electricity price, and the revenue obtained is not as good as selling part of the predicted power in the real-time market. Therefore, in order to obtain more revenue, it will be chosen to reduce it on the basis of the predicted power, hoping to sell the reduced part of the power in the real-time market at the real-time electricity price. In this step, the power value obtained by reducing the predicted power of the target time is the day-ahead declared power of the target time.

[0066] S112, if the probability of negative price difference is greater than the first threshold, then the predicted power corresponding to the target time is increased to obtain the day-ahead declared power corresponding to the target time.

[0067] It can be understood that the first threshold is a pre-set judgment standard value. If the probability of a negative price difference is greater than the first threshold, it means that the probability of a negative price difference is greater than that of a positive price difference, which also means that the possibility of a day-ahead electricity price being higher than the real-time electricity price at the target time is greater. Overreporting the predicted power appropriately will result in more power being sold at the day-ahead electricity price, and the excess can be purchased and supplemented at a lower real-time electricity price in the real-time market to achieve benefits. Therefore, in order to obtain more profits, it will be chosen to increase the predicted power, hoping to obtain the price difference between the day-ahead market and the real-time market for the increased power. In this step, the power value obtained by increasing the predicted power of the target time is the day-ahead declared power of the target time.

[0068] This method selects the most valuable similar historical days from multiple historical days within a preset time span based on the similarity of load information, and takes the positive price difference probability and negative price difference probability of similar historical days as a reference, and reduces or increases the predicted power of each target time period to obtain the day-ahead declared power of each target time period. This method can automatically determine the day-ahead declared power after the power trading center releases the day-ahead information, providing a scientific basis for market participants to make decisions, reducing declaration risks, and increasing the possibility of benefits. It also improves the work efficiency of distributed photovoltaic aggregators and reduces the workload caused by manual accounting.

[0069] In one embodiment, the historical load information includes a variety of other historical loads that affect the market-based load, and the day-ahead load information includes a variety of other day-ahead loads that affect the market-based load. The other historical loads here are of the same type as the other day-ahead loads, including centralized loads, non-market-based loads, market-based loads, etc. In short, the centralized load itself and its various components have an impact on the size of the market-based load, and the similarity of the loads on two trading days can be judged from the similarity between the various loads other than the market-based load.

[0070] Please refer to Figure 2 , matching the load information published on the target trading day before the target time-sharing day with the historical load information on each historical day at the target time-sharing day, and selecting the historical days corresponding to the most similar preset number of historical load information as similar historical days, including steps S202 to S204.

[0071] S202, for any historical day, determine the first similarity between the historical market-based load at the target time and the market-based load of the previous day on the historical day, and then determine the second similarities between other historical loads and other previous day loads of the same type, and obtain the third similarity corresponding to the historical day based on the first similarity, each second similarity and its corresponding first weight; the first weight corresponding to the first similarity is the largest, and the first weight corresponding to the second similarity obtained based on the type with a greater impact on the market-based load is larger.

[0072] S204: Select a preset number of historical days as similar historical days in descending order of the third similarity.

[0073] It can be understood that the amount of market-oriented load is the most important factor in determining the electricity price. The first similarity can reflect the similarity of the market-oriented load between the historical day and the target trading day. Therefore, it can largely reflect the electricity price situation of the two days at the same time. The first similarity can select the market-oriented load of the target trading day as the base number, compare the historical market-oriented load of each historical day with it, and then normalize the results to obtain the first similarity. There are many other loads that affect the market-oriented load. By comparing the specific data of the same other load on the historical day and on the target trading day, a second similarity corresponding to another load can be obtained. For example, by comparing the historical unified load of the historical day with the unified load of the target trading day, the second similarity corresponding to the unified load can be obtained. By comparing the historical non-market load of the historical day with the non-market load of the target trading day, the second similarity corresponding to the non-market load can be obtained. The specific method of comparison can be to divide each historical data with the data of the day before, and then normalize the results of each division, and use the normalized result as the second similarity.

[0074] In addition, since the market-oriented load itself has the highest impact on electricity prices, its corresponding first weight is the largest, while other loads have different impacts on the market-oriented load. The type with the greater impact on the market-oriented load has a higher importance of the corresponding second similarity when comprehensively considered, so the first weight assigned to it is greater. For example, the impact of the centralized load on the market-oriented load is greater than that on the non-market-oriented load, so the first weight corresponding to the centralized load will also be greater. By weighted summing each second similarity according to the corresponding first weight, the third similarity that combines various influencing factors can be obtained.

[0075] The third similarity well reflects the similarity between the load conditions of the historical day and the target trading day. After obtaining the third similarity corresponding to each historical day, the historical days are arranged in descending order according to the third similarity. In this way, the historical day that is most similar to the target trading day can be arranged in front, and then a preset number of historical days can be selected as similar historical days.

[0076] In one embodiment, see Figure 3 If the positive price difference probability is greater than the first threshold, it is reduced on the basis of the predicted power corresponding to the target time-sharing to obtain the day-ahead declared power corresponding to the target time-sharing, including steps S302 to S306.

[0077] S302: When the positive price spread probability is greater than a first threshold, determine a first target interval to which the positive price spread probability belongs among a plurality of consecutive first intervals.

[0078] It can be understood that in order to determine the adjustment range more precisely, the value range of the positive spread probability above the first threshold is divided into multiple continuous first intervals. In this way, different degrees of adjustment measures can be taken according to different probability ranges. Each first interval represents a positive spread probability value within a certain range. By judging the position of the positive spread probability in these intervals, the approximate range of the corresponding adjustment strategy can be determined. For example, if the first threshold is selected as 0.5, the value range of the positive spread probability from 0.5 to 1 can be divided into several intervals, such as (0.5, 0.65], (0.65, 0.8], (0.8, 0.9], (0.9, 1]. After the positive spread probability is determined, by comparing its value with the range of each interval, the first target interval to which it belongs can be determined. If the positive spread probability is 0.75, then it belongs to the interval (0.65, 0.8], and this interval is the first target interval.

[0079] S304, determining a target reduction according to the first mapping relationship and the first target interval. The first mapping relationship reflects a one-to-one correspondence between each first interval and an optional reduction. The larger the probability value included in the first interval, the larger the corresponding optional reduction.

[0080] It can be understood that in order to more scientifically determine the reduction when the probability of positive price difference is greater than the first threshold, a first mapping relationship is established. This mapping relationship reflects the one-to-one correspondence between each first interval and the optional reduction. That is to say, for each first interval, there is a specific optional reduction corresponding to it. Moreover, the larger the probability value contained in the first interval, the larger the corresponding optional reduction. This is because when the probability of positive price difference is large, it means that the possibility that the real-time electricity price is higher than the electricity price reported a day ago is very high, and it is necessary to take larger adjustment measures to increase the revenue. For example, if the first interval (0.5,0.65] corresponds to an optional reduction of 10%, (0.65, 0.8] corresponds to an optional reduction of 20%, [0.8, 0.9) corresponds to an optional reduction of 30%, and (0.9, 1] corresponds to an optional reduction of 40%. After determining the first target interval to which the probability of positive price difference belongs, the corresponding optional reduction can be found as the target reduction according to the first mapping relationship.

[0081] S306, reducing the predicted power according to the target reduction range to obtain the day-ahead power reported corresponding to the target time period.

[0082] It can be understood that the specific calculation method can be to multiply the predicted power by , get the day-ahead power. For example, if the forecast power is 100 MW and the target reduction is 30%, then the day-ahead power is .

[0083] In one embodiment, see Figure 4 If the probability of negative price difference is greater than the first threshold, the predicted power corresponding to the target time-sharing is increased to obtain the day-ahead declared power corresponding to the target time-sharing, including steps S402 to S406.

[0084] S402: When the negative price difference probability is greater than the first threshold, determine the second target interval to which the negative price difference probability belongs among a plurality of consecutive second intervals.

[0085] It can be understood that in order to determine the adjustment range more precisely, the value range of the negative price difference probability above the first threshold is divided into multiple continuous second intervals. In this way, different degrees of adjustment measures can be taken according to different probability ranges. Each second interval represents a negative price difference probability value within a certain range. By judging the position of the negative price difference probability in these intervals, the approximate range of the corresponding adjustment strategy can be determined. For example, if the first threshold is selected as 0.5, the value range of the negative price difference probability from 0.5 to 1 can be divided into several intervals, such as (0.5, 0.65], (0.65, 0.8], (0.8, 0.9], (0.9, 1]. After the negative price difference probability is determined, by comparing its value with the range of each interval, the second target interval to which it belongs can be determined. If the negative price difference probability is 0.75, then it belongs to the interval (0.65, 0.8], and this interval is the second target interval.

[0086] S404, determining a target increase according to the second mapping relationship and the second target interval. The second mapping relationship reflects a one-to-one correspondence between each second interval and an optional increase. The larger the probability value contained in the second interval, the larger the corresponding optional increase.

[0087] It can be understood that in order to more scientifically determine the increase when the probability of a negative price difference is greater than the first threshold, a second mapping relationship is established. This mapping relationship reflects the one-to-one correspondence between each second interval and the optional increase. That is to say, for each second interval, there is a specific optional increase corresponding to it. Moreover, the larger the probability value contained in the second interval, the larger the corresponding optional increase. This is because when the probability of a negative price difference is large, it means that the possibility that the electricity price declared a day ago is higher than the real-time electricity price is very high, and it is necessary to take larger adjustment measures to increase revenue. For example, if the optional increase corresponding to the second interval (0.5, 0.65] is 10%, the optional increase corresponding to (0.65, 0.8] is 20%, the optional increase corresponding to [0.8, 0.9] is 30%, and the optional increase corresponding to (0.9, 1] is 40%. After determining the second target interval to which the probability of a negative price difference belongs, the corresponding optional increase can be found as the target increase according to the second mapping relationship.

[0088] S406, increasing the predicted power according to the target increase, and obtaining the day-ahead power corresponding to the target time.

[0089] It can be understood that the specific calculation method can be to multiply the predicted power by (1 + target increase) to obtain the day-ahead declared power. For example, if the predicted power is 100 MW and the target increase is 30%, then the day-ahead declared power is 100×(1 +0.3) = 130 MW.

[0090] In one embodiment, the predicted power at each target time on the target trading day is determined, including: performing power prediction based on multiple weather forecast data of the target time and its corresponding second weight to obtain the predicted power at the target time. It can be understood that photovoltaic power generation will be greatly affected by the weather. When performing power prediction, it is necessary to consider multiple weather forecast data, but each weather forecast data has a different degree of influence on the size of photovoltaic power generation. For this reason, corresponding weights are set for various weather forecast data in the relevant power prediction algorithm, that is, the second weight in this embodiment.

[0091] In one embodiment, see Figure 5 , power prediction is performed according to the multiple meteorological forecast data of the target time and the corresponding second weights to obtain the predicted power of the target time, including steps S502 to S506.

[0092] S502, determining the forecast accuracy of various weather forecast data within a preset time span.

[0093] It can be understood that forecast accuracy is an indicator that quantifies the degree of conformity between weather forecast data and actual weather conditions. For each type of weather forecast data, whether the difference between the forecast data and the actual value exceeds the threshold corresponding to the weather forecast data can be used to determine whether the forecast data is accurate. The forecast accuracy can be obtained by dividing the number of accurate days by the total number of historical days.

[0094] S504, for any kind of weather forecast data, determine the adjustment coefficient according to the forecast accuracy and the third mapping relationship, multiply the target adjustment coefficient by the initial weight corresponding to the weather forecast data to obtain the second weight corresponding to the weather forecast data.

[0095] It can be understood that in various photovoltaic power prediction algorithms, it is often assumed that the weather forecast is accurate, and the weights of various weather forecast data are set under such ideal conditions. The weight is the initial weight, and the initial weight can be directly used as the second weight. However, in actual situations, there is still a probability that the weather forecast data will be biased. When predicting power, more emphasis should be placed on referring to data with high credibility. Therefore, the weather forecast data with a smaller probability of error should have an appropriately increased weight in power prediction. For this reason. The present embodiment pre-sets a third mapping relationship, which reflects the relationship between the forecast accuracy and the adjustment coefficient. The adjustment coefficient is less than or equal to 1, and the higher the forecast accuracy, the closer the corresponding adjustment coefficient is to 1. This means that if the accuracy of a type of weather forecast data is very high, then its adjustment coefficient will be close to 1, and the adjustment to its initial weight will be smaller; if the forecast accuracy is low, the adjustment coefficient will be less than 1, and the adjustment to its initial weight will be larger.

[0096] The contribution of different meteorological forecast data to power prediction may vary depending on the forecast accuracy. Therefore, it is necessary to adjust the weights of various meteorological forecast data according to the forecast accuracy. First, the adjustment coefficient is determined according to the forecast accuracy through the third mapping relationship. This mapping relationship reflects the relationship between the forecast accuracy and the adjustment coefficient. The adjustment coefficient is less than or equal to 1, and the higher the forecast accuracy, the closer the corresponding adjustment coefficient is to 1. This means that if the accuracy of a kind of meteorological forecast data is very high, then its adjustment coefficient will be close to 1, and the adjustment of its initial weight will be smaller; if the forecast accuracy is low, the adjustment coefficient will be less than 1, and the adjustment of its initial weight will be larger. Then, the target adjustment coefficient is multiplied by the initial weight corresponding to the meteorological forecast data to obtain the second weight corresponding to the meteorological forecast data. In this way, the weights of various meteorological forecast data can be dynamically adjusted according to the forecast accuracy, so that meteorological forecast data with higher accuracy is given more attention in power prediction. The third mapping relationship is preferably a nonlinear mapping, with the forecast accuracy as the horizontal axis and the adjustment coefficient as the vertical axis. The closer the entire curve is to the high forecast accuracy, the smaller the slope is, and the closer the slope is to the low forecast accuracy, the larger the slope is. That is, as the forecast accuracy decreases, the rate of decrease of the adjustment coefficient will gradually increase.

[0097] S506, performing power prediction according to the various meteorological forecast data of the target time and the corresponding second weights to obtain the predicted power of the target time.

[0098] The present application provides a device for determining the day-ahead power reported by a distributed photovoltaic aggregator, including a data acquisition module, a power prediction module, a matching module, a probability determination module, a first adjustment module, and a second adjustment module.

[0099] The data acquisition module is used to obtain the historical electricity price information and historical load information corresponding to each time-sharing period on multiple historical days within a preset time span. The historical electricity price information includes the historical day-ahead electricity price and the historical real-time electricity price. The power prediction module is used to determine each time-sharing period as the target time-sharing period, and determine the predicted power at each target time-sharing period on the target trading day. The matching module is used to match the load information released on the day-ahead of the target trading day with the historical load information of each historical day at the target time-sharing period, and select the historical days corresponding to the most similar preset number of historical load information as the similar historical days corresponding to the target time-sharing period. The probability determination module is used to determine the positive price difference probability and negative price difference probability of the target time-sharing period according to the historical electricity price information corresponding to the target time-sharing period on each similar historical day corresponding to the target time-sharing period. The first adjustment module is used to reduce the predicted power corresponding to the target time-sharing period when the positive price difference probability is greater than the first threshold, and obtain the day-ahead declared power corresponding to the target time-sharing period. The second adjustment module is used to increase the predicted power corresponding to the target time-sharing period when the negative price difference probability is greater than the first threshold, and obtain the day-ahead declared power corresponding to the target time-sharing period.

[0100] For the specific limitations of the device for determining the power reported by distributed photovoltaic aggregators the day before, please refer to the limitations of the method for determining the power reported by distributed photovoltaic aggregators the day before, and no further details will be given here. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. It should be noted that the division of modules in the embodiment of the present application is schematic, which is only a logical function division, and there may be other division methods in actual implementation.

[0101] The present application provides a computer device, including one or more processors and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the one or more processors, the steps of the method for determining the power reported by the distributed photovoltaic aggregator on the previous day in any of the above-mentioned embodiments are executed.

[0102] Indicatively, Figure 6 As shown, Figure 6 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. Figure 6The computer device 600 includes a processing component 602, which further includes one or more processors, and a memory resource represented by a memory 601, for storing instructions executable by the processing component 602, such as an application. The application stored in the memory 601 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 602 is configured to execute instructions to perform the steps of the method for determining the power reported by the distributed photovoltaic aggregator in any of the above embodiments.

[0103] The computer device 600 may further include a power supply component 603 configured to perform power management of the computer device 600 , a wired or wireless network interface 604 configured to connect the computer device 600 to a network, and an input / output (I / O) interface 605 .

[0104] Those skilled in the art will understand that Figure 6 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.

[0105] The present application provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the method for determining the power reported by a distributed photovoltaic aggregator on the previous day in any of the above-mentioned embodiments.

[0106] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can refer to each other.

[0107] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for determining the power reported by a distributed photovoltaic aggregator on the previous day, characterized in that: include: Obtain historical electricity price information and historical load information corresponding to each time-sharing period for multiple historical days within a preset time span; the historical electricity price information includes historical day-ahead electricity prices and historical real-time electricity prices; Determine each time-sharing as a target time-sharing, and determine the predicted power at each target time-sharing on the target trading day; Match the load information published on the target trading day before the target time-sharing day with the historical load information on each historical day at the target time-sharing day, and select the historical days corresponding to the most similar preset number of historical load information as similar historical days corresponding to the target time-sharing day; Determine the positive price difference probability and the negative price difference probability of the target time-sharing according to the historical electricity price information corresponding to the target time-sharing on each of the similar historical days corresponding to the target time-sharing; If the positive price difference probability is greater than the first threshold, the predicted power corresponding to the target time is reduced to obtain the day-ahead declared power corresponding to the target time; If the negative price difference probability is greater than the first threshold, it is increased based on the predicted power corresponding to the target time-sharing to obtain the day-ahead declared power corresponding to the target time-sharing.

2. The method for determining the power reported by the distributed photovoltaic aggregator on the previous day according to claim 1 is characterized in that: The historical load information includes a variety of other historical loads that affect the historical market-based loads, and the day-ahead load information includes a variety of other day-ahead loads that affect the day-ahead market-based loads; The step of matching the load information published on the target trading day before the target time-sharing day with the historical load information on each historical day at the target time-sharing day, and selecting the historical days corresponding to the most similar preset number of historical load information as similar historical days, includes: For any one of the historical days, determine the first similarity between the historical market-based load and the day-ahead market-based load at the target time-sharing time of the historical day, and then respectively determine the second similarities between other historical loads and other day-ahead loads of the same type, and obtain the third similarity corresponding to the historical day according to the first similarity, each of the second similarities and their corresponding first weights; the first weight corresponding to the first similarity is the largest, and the first weight corresponding to the second similarity obtained based on the type with the greater impact on the market-based load is greater; According to the order of the third similarity from high to low, the preset number of historical days are selected as the similar historical days.

3. The method for determining the power reported by the distributed photovoltaic aggregator on the previous day according to claim 1 is characterized in that: If the positive price difference probability is greater than the first threshold, the predicted power corresponding to the target time is reduced to obtain the day-ahead declared power corresponding to the target time, including: When the positive price difference probability is greater than the first threshold, determining a first target interval to which the positive price difference probability belongs among a plurality of consecutive first intervals; Determine a target reduction according to a first mapping relationship and the first target interval; the first mapping relationship reflects a one-to-one correspondence between each of the first intervals and an optional reduction, and the larger the probability value included in the first interval, the larger the corresponding optional reduction; According to the target reduction rate, the predicted power is reduced to obtain the day-ahead declared power corresponding to the target time-sharing period.

4. The method for determining the power reported by the distributed photovoltaic aggregator on the previous day according to claim 1 is characterized in that: If the negative price difference probability is greater than the first threshold, the predicted power corresponding to the target time is increased to obtain the day-ahead declared power corresponding to the target time, including: When the negative price difference probability is greater than the first threshold, determining a second target interval to which the negative price difference probability belongs among a plurality of consecutive second intervals; Determine a target increase according to a second mapping relationship and the second target interval; the second mapping relationship reflects a one-to-one correspondence between each of the second intervals and the optional increase, and the larger the probability value contained in the second interval, the larger the corresponding optional increase; According to the target increase, the predicted power is increased to obtain the day-ahead power corresponding to the target time.

5. The method for determining the power reported by the distributed photovoltaic aggregator on the previous day according to claim 1 is characterized in that: Determining the predicted power at each target time on the target trading day includes: Power prediction is performed according to the multiple meteorological forecast data of the target time and the corresponding second weights to obtain the predicted power of the target time.

6. The method for determining the power reported by the distributed photovoltaic aggregator on the previous day according to claim 5 is characterized in that: The power prediction is performed according to the multiple meteorological forecast data of the target time and the corresponding second weights to obtain the predicted power of the target time, including: Determining the forecast accuracy of various weather forecast data within the preset time span; For any type of the weather forecast data, an adjustment coefficient is determined according to the forecast accuracy and the third mapping relationship, and the target adjustment coefficient is multiplied by the initial weight corresponding to the weather forecast data to obtain the second weight corresponding to the weather forecast data; the third mapping relationship reflects the mapping relationship between the forecast accuracy and the adjustment coefficient, the adjustment coefficient is less than or equal to 1, and the higher the forecast accuracy, the closer the corresponding adjustment coefficient is to 1; Power prediction is performed according to the multiple meteorological forecast data of the target time and the second weights corresponding thereto to obtain the predicted power of the target time.

7. The method for determining the power reported by the distributed photovoltaic aggregator on the previous day according to claim 1 is characterized in that: Determining the positive price difference probability and the negative price difference probability of the target time-sharing according to the historical electricity price information corresponding to the target time-sharing on each of the similar historical days corresponding to the target time-sharing includes: If the historical real-time electricity price of the similar historical day at the target time is higher than the historical day-ahead electricity price, the similar historical day is classified as a positive price difference, otherwise, it is classified as a negative price difference; Obtaining the positive price difference probability according to a ratio of the number of the historical similar days of the positive price difference category to the preset number; The negative price difference probability is obtained according to the ratio of the number of the historical similar days of the category of the negative price difference to the preset number.

8. A device for determining the power reported by a distributed photovoltaic aggregator on the previous day, characterized in that: include: A data acquisition module is used to obtain historical electricity price information and historical load information corresponding to each time period for multiple historical days within a preset time span; the historical electricity price information includes historical day-ahead electricity prices and historical real-time electricity prices; A power prediction module, used to determine each time-sharing as a target time-sharing, and determine the predicted power at each target time-sharing on the target trading day; A matching module, for matching the load information published on the target trading day before the target time-sharing day with the historical load information on each historical day at the target time-sharing day, and selecting the historical days corresponding to the most similar preset number of historical load information as similar historical days corresponding to the target time-sharing day; A probability determination module, for determining a positive price difference probability and a negative price difference probability of the target time-sharing according to the historical electricity price information corresponding to the target time-sharing on each of the similar historical days corresponding to the target time-sharing; A first adjustment module is used for reducing the predicted power corresponding to the target time-sharing to obtain the day-ahead declared power corresponding to the target time-sharing when the positive price difference probability is greater than a first threshold; The second adjustment module is used to increase the predicted power corresponding to the target time-sharing when the probability of the negative price difference is greater than the first threshold, so as to obtain the day-ahead declared power corresponding to the target time-sharing.

9. A computer device, characterized in that: It includes one or more processors and a memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the one or more processors, the steps of the method for determining the power reported by the distributed photovoltaic aggregator in the previous day as described in any one of claims 1 to 7 are performed.

10. A storage medium, characterized in that: The storage medium stores computer-readable instructions, and when the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the method for determining the power reported by a distributed photovoltaic aggregator as described in any one of claims 1-7.

Images