Load baseline prediction method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a load baseline prediction method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring historical load data before and after a demand response day, and extracting historical load characteristics according to the historical load data; acquiring a corresponding load feature vector according to the historical load feature; screening typical days meeting a preset condition as a first typical day set, and extracting a load feature vector corresponding to each typical day; based on different scenes, matching the corresponding load feature vectors with the corresponding load feature vector of each typical day, and screening out the successfully matched typical days in the first typical day set as a second typical day set; and performing prediction according to the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain a load baseline of the demand response day to be predicted. Therefore, the prediction mode better fits a specific scene, and the prediction result is more accurate.

Description

Technical Field

[0001] The present application relates to the technical field of power demand response, and in particular to a load baseline prediction method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] Traditionally, residential loads participate in demand response in an aggregated manner after being represented by load merchants. When regional electricity demand peaks, the power grid company issues load reduction tasks to load aggregators, who break down the tasks and push information such as control requirements to residents. Residents reduce their own electricity demand by increasing the air outlet temperature of air conditioners, postponing laundry, or adjusting the refrigerator cooling mode. The load reduction task of the power grid is finally completed by superimposing the regulation effects of a large number of residential loads. By calculating the baseline of residential loads, the effect of demand response can be accurately evaluated and a reasonable electricity price strategy can be formulated.

[0003] The existing technology mainly considers how to improve the accuracy of load baseline calculation and correct the baseline load size through algorithms. However, baseline load is not suitable for residential load in practical applications for the following reasons: First, there are a large number of residents. Calculating the load adjustment of residents is not only a large workload for aggregators, but also cumbersome. It is necessary to call for data from the meter and conduct unified analysis at the master station, which is difficult to complete in a short time; second, if the residential load is measured in the form of baseline measurement, the contribution of the demand response will be affected by environmental factors, and there will be errors, omissions, and distortions in the data measurement, making it difficult to verify and complete the data. Summary of the invention

[0004] Based on this, it is necessary to provide a load baseline prediction method, device, computer equipment, computer-readable storage medium and computer program product suitable for residential load in response to the above technical problems.

[0005] In a first aspect, the present application provides a load baseline prediction method, comprising:

[0006] Obtain historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data;

[0007] According to the historical load characteristics, a load characteristic vector corresponding to the demand response day that has occurred is obtained;

[0008] Selecting typical days that meet preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set;

[0009] Based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and the typical days that are successfully matched in the first typical day set are selected as the second typical day set;

[0010] Prediction is performed based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

[0011] In one embodiment, the load characteristic includes at least one of the maximum load, the minimum load, the average load, the peak-to-valley difference rate or the average change rate in the load data.

[0012] In one embodiment, the step of selecting typical days that meet preset conditions as the first typical day set includes:

[0013] Screening typical days that are in the same season as the demand response days that have occurred to obtain a first data set;

[0014] Screening out typical days with the same classification type as the demand response day that has occurred from the first data set to obtain a second data set, wherein the classification type includes working days and non-working days;

[0015] From the second data set, typical days whose similarity with the historical load data of the demand response day that has occurred reaches a preset degree are screened out to obtain a first typical day set.

[0016] In one embodiment, based on different scenarios, matching the load characteristic vector corresponding to the demand response day that has occurred with the load characteristic vector corresponding to each typical day in the first typical day set, and selecting the typical days in the first typical day set that are successfully matched, includes:

[0017] Determine the load scenario corresponding to the demand response day that has occurred based on whether the load that needs to be adjusted on the demand response day that has occurred includes transferable load;

[0018] According to the load scenario corresponding to the demand response day that has occurred, a load characteristic vector corresponding to the demand response day that has occurred is obtained;

[0019] Calculate the cosine similarity between the load characteristic vector corresponding to the demand response day that has occurred and the load characteristic vector corresponding to each typical day in the first typical day set;

[0020] The typical day whose cosine similarity is greater than a preset value is taken as the typical day successfully matched in the first typical day set.

[0021] In one embodiment, determining the load scenario corresponding to the demand response day that has occurred according to whether the load to be adjusted on the demand response day that has occurred includes a transferable load includes:

[0022] Obtaining a first time curve and a second time curve, wherein the first time curve is a time curve formed by historical load data around the demand response day that has occurred, in the case where demand response is performed on the demand response day that has occurred, and the second time curve is a time curve formed by historical load data around the demand response day that has occurred, in the case where demand response is not performed on the demand response day that has occurred;

[0023] For the first partial curve corresponding to the first time curve on the adjacent day after the demand response day that has occurred, and the second partial curve corresponding to the second time curve on the adjacent day after the demand response day that has occurred, if the change trends of the first partial curve and the second partial curve are the same, the load that needs to be adjusted on the demand response day that has occurred does not include the transferable load, and the load scenario corresponding to the demand response day that has occurred is the first load scenario; if the change trends of the first partial curve and the second partial curve are different, the load that needs to be adjusted on the demand response day that has occurred includes the transferable load, and the load scenario corresponding to the demand response day that has occurred is the second load scenario.

[0024] In one embodiment, the method further comprises:

[0025] Calculating a cluster center in the second typical day set, where the cluster center is a typical day in the second typical day set;

[0026] Calculating a preset range of a load baseline based on historical load data in the cluster center;

[0027] Determine whether the load baseline of each time period within the demand response day to be predicted is within a preset range. If it is within the preset range, the load baseline of the corresponding time period remains unchanged; if it is not within the preset range, the load data of the corresponding time period in the cluster center is used to replace the load baseline of the corresponding time period.

[0028] In a second aspect, the present application also provides a load baseline prediction device, comprising:

[0029] An acquisition module is used to acquire historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data;

[0030] The acquisition module is further used to acquire a load characteristic vector corresponding to a demand response day that has occurred according to the historical load characteristics;

[0031] A screening module, used for screening typical days meeting preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set;

[0032] The screening module is further used to match the load characteristic vector corresponding to the demand response day that has occurred with the load characteristic vector corresponding to each typical day in the first typical day set based on different scenarios, and screen out the typical days in the first typical day set that are successfully matched as the second typical day set;

[0033] The prediction module is used to make a prediction based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

[0034] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0035] Obtain historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data;

[0036] According to the historical load characteristics, a load characteristic vector corresponding to the demand response day that has occurred is obtained;

[0037] Selecting typical days that meet preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set;

[0038] Based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and the typical days that are successfully matched in the first typical day set are selected as the second typical day set;

[0039] Prediction is performed based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

[0040] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0041] Obtain historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data;

[0042] According to the historical load characteristics, a load characteristic vector corresponding to the demand response day that has occurred is obtained;

[0043] Selecting typical days that meet preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set;

[0044] Based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and the typical days that are successfully matched in the first typical day set are selected as the second typical day set;

[0045] Prediction is performed based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

[0046] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0047] Obtain historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data;

[0048] According to the historical load characteristics, a load characteristic vector corresponding to the demand response day that has occurred is obtained;

[0049] Selecting typical days that meet preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set;

[0050] Based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and the typical days that are successfully matched in the first typical day set are selected as the second typical day set;

[0051] Prediction is performed based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

[0052] The load baseline prediction method, device, computer equipment, computer-readable storage medium and computer program product obtain the historical load data before and on the demand response day that has occurred, extract the historical load characteristics before and on the demand response day based on the historical load data; obtain the load characteristic vector corresponding to the demand response day that has occurred based on the historical load characteristics; select the typical days that meet the preset conditions as the first typical day set, and extract the load characteristic vector corresponding to each typical day in the first typical day set; based on different scenarios, match the load characteristic vector corresponding to the demand response day that has occurred with the load characteristic vector corresponding to each typical day in the first typical day set, and select the typical days that successfully match in the first typical day set as the second typical day set; predict based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set, and obtain the load baseline of the demand response day to be predicted. In this way, by fully considering the load data before and after the demand response day that has occurred, extracting the load characteristics of the load data, taking the load characteristics as the basis for matching with the typical day, and selecting the typical day that meets the conditions. Combined with the characteristics of residents' electricity consumption, the demand response day is divided into two specific scenarios. Under the specific scenarios, the residents' load baseline is predicted, making the prediction results of the load baseline more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] 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.

[0054] FIG1 is a schematic diagram of a flow chart of a load baseline prediction method in one embodiment;

[0055] FIG2 is a schematic flow chart of a load baseline prediction method in another embodiment;

[0056] FIG3 is a graph of load data versus time in one embodiment;

[0057] FIG4 is a block diagram of a load baseline prediction device in one embodiment;

[0058] FIG. 5 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0059] 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.

[0060] The load baseline prediction method provided in the embodiment of the present application can be applied to the scenario where residential electricity consumption needs to respond to demand during peak electricity demand, and the load baseline of residential electricity consumption can be estimated in advance, so as to accurately evaluate the effect of residential electricity consumption responding to demand and formulate a reasonable subsidy strategy. Among them, demand response is a series of load reduction behaviors performed by residents according to requirements.

[0061] In one embodiment, as shown in FIG1 , a load baseline prediction method is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0062] Step 102, obtaining historical load data before and on the demand response day that has occurred, and extracting historical load characteristics before and on the demand response day based on the historical load data.

[0063] Among them, load data refers to the size of power consumption in each time period and the corresponding time; load characteristics refer to at least one of the maximum load, minimum load, average load, peak-to-valley difference rate or average change rate in the load data.

[0064] Specifically, the terminal will obtain historical load data before and on the demand response day that has occurred, and extract at least one of the maximum load, minimum load, average load, peak-to-valley difference rate or average change rate in the historical load data.

[0065] Step 104: Obtain a load characteristic vector corresponding to a demand response day that has occurred based on historical load characteristics.

[0066] In some embodiments, the terminal obtains a load characteristic vector corresponding to a demand response day that has occurred based on historical load characteristics, wherein the load characteristic vector corresponding to a demand response day that has occurred is composed of at least one of the maximum load, minimum load, average load, peak-to-valley difference rate, or average change rate in the historical load data.

[0067] Step 106 , selecting typical days that meet preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set.

[0068] Specifically, the terminal will select typical days that meet the preset conditions as the first typical day set, and extract at least one of the maximum load, minimum load, average load, peak-to-valley difference rate or average change rate from the load data corresponding to each typical day in the first typical day set, so as to further obtain the load characteristic vector corresponding to each typical day in the first typical day set.

[0069] Step 108, based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and the typical days that are successfully matched in the first typical day set are selected as the second typical day set.

[0070] In some embodiments, the terminal will divide the demand response day that has occurred into two specific load scenarios based on the characteristics of residential electricity consumption, and calculate the cosine similarity between the load characteristic vector corresponding to the demand response day that has occurred and the load characteristic vector corresponding to each typical day in the first typical day set in different scenarios, and take the typical day in the first typical day set whose cosine similarity is greater than a preset value as the second typical day set.

[0071] Step 110 , forecasting is performed based on the historical load data before the demand response day to be forecasted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be forecasted.

[0072] In some embodiments, the terminal uses a prediction algorithm to predict the load baseline after the demand response day to be predicted based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set. The prediction algorithm can be any algorithm that can predict the result, and this is not limited in this embodiment. The number of typical days in the second typical day set is not less than 1.

[0073] Specifically, the prediction algorithm formula is shown in (1).

[0074]

[0075] Where f(·) is the prediction algorithm, Y is the load data set of I second typical days, and the specific load data is expressed as shown in formula (2).

[0076]

[0077] is the load data set from time period 1 to T on a typical day i, X is the historical load data set for each time period before the demand response is executed on the demand response day, , the baseline load on the demand response day obtained by the prediction algorithm is , is the baseline load set on the demand response day, is the baseline load size during the demand response day period t, represents the time when demand response starts, Indicates the time when the demand response ends. Represents the total number of time periods in the entire process.

[0078] In the above load baseline prediction method, first, the historical load data before and on the demand response day that has occurred are obtained, and the historical load characteristics before and on the demand response day that have occurred are extracted based on the historical load data; based on the historical load characteristics, the load characteristic vector corresponding to the demand response day that has occurred is obtained; the typical days that meet the preset conditions are selected as the first typical day set, and the load characteristic vector corresponding to each typical day in the first typical day set is extracted; based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and the typical days that successfully match in the first typical day set are selected as the second typical day set; the load baseline of the demand response day to be predicted is obtained based on the historical load data of each period before the demand response is executed on the demand response day to be predicted and the historical load data of each typical day in the second typical day set. In this way, by fully considering the load data before and after the demand response day that has occurred, the load characteristics of the load data are extracted, and the load characteristics are used as the basis for matching with the typical day, and the typical days that meet the conditions are selected. Combined with the characteristics of residents' electricity consumption, the demand response day is divided into two specific scenarios. Under the specific scenarios, the residents' load baseline is predicted, making the prediction results of the load baseline more accurate.

[0079] In an exemplary embodiment, the load characteristic includes at least one of a maximum load, a minimum load, an average load, a peak-to-valley difference rate, or an average change rate in the load data.

[0080] In actual implementation, the load characteristic includes at least one of the maximum load, the minimum load, the average load, the peak-to-valley difference rate or the average change rate in the load data.

[0081] Furthermore, the specific calculation formula for extracting the historical load characteristics of each time period before the demand response is executed on the demand response day from the historical load data is shown in (3), and the specific calculation formula for extracting the historical load characteristics of each time period on the day after the demand response is executed on the demand response day from the historical load data is shown in (4).

[0082]

[0083]

[0084] in, represents the maximum load on the day of demand response, represents the minimum load on the day before the demand response, represents the average load on the day before the demand response, represents the average rate of change of demand response day before, It represents the peak-to-valley difference rate of the demand response day. represents the maximum load after the demand response day, represents the minimum load after the demand response day, represents the average load after the demand response day, represents the average rate of change of demand response days later, It represents the peak-to-valley difference rate after the demand response day. represents the load on the day before the demand response, represents the time when demand response starts, Indicates the time when the demand response ends. Indicates the total number of time periods in a day.

[0085] Furthermore, the specific calculation formula of the load characteristic vector corresponding to the demand response day that has occurred is shown in formula (5).

[0086]

[0087] In the above embodiment, the load characteristics are represented by using at least one of the maximum load, minimum load, average load, peak-to-valley difference rate or average change rate in the load data, and the characteristics of the load data are quantified, so that the matching results of subsequent typical days are more accurate.

[0088] In some exemplary embodiments, typical days that meet preset conditions are screened as a first typical day set, including: screening typical days that are in the same season as the demand response days that have occurred, to obtain a first data set; screening out typical days with the same classification type as the demand response days that have occurred from the first data set, to obtain a second data set, wherein the classification types include working days and non-working days; from the second data set, screening out typical days whose degree of similarity with the historical load data of the demand response days that have occurred reaches a preset degree, to obtain a first typical day set.

[0089] In actual implementation, the terminal will first determine the season of the demand response day that has occurred, and filter out typical days in the same season as the demand response day that has occurred to obtain the first data set; determine the classification type of the demand response day that has occurred. If it is a working day, the typical days that are working days in the first data set are filtered out as the second data set; if the demand response day that has occurred is a non-working day, the typical days that are non-working days in the first data set are filtered out as the second data set.

[0090] Further, the electricity consumption data of each typical day in the second data set and the electricity consumption data of the demand response day that has occurred when the demand response is not executed are obtained, and the typical days in the second data set whose difference between the electricity consumption data of each typical day in the second data set and the electricity consumption data of the demand response day that has occurred when the demand response is not executed is less than a preset range are screened out as the first typical day set. The preset range is determined by the terminal according to actual conditions, and this embodiment does not limit this. The number of typical days in the first typical day set is not less than n.

[0091] In the above embodiment, typical days similar to the demand response day are screened step by step, and the baseline load after the demand response day to be predicted is predicted using the data of the screened typical days, so that the obtained result is more accurate.

[0092] In some exemplary embodiments, based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and typical days that successfully match the first typical day set are screened out, including: determining the load scenario corresponding to the demand response day that has occurred according to whether the load to be adjusted on the demand response day that has occurred includes transferable load; obtaining the load characteristic vector corresponding to the demand response day that has occurred according to the load scenario corresponding to the demand response day that has occurred; calculating the cosine similarity between the load characteristic vector corresponding to the demand response day that has occurred and the load characteristic vector corresponding to each typical day in the first typical day set; and taking the typical day with a cosine similarity greater than a preset value as a typical day that successfully matches the first typical day set.

[0093] In some exemplary embodiments, according to whether the load that needs to be adjusted on the demand response day that has occurred includes a transferable load, determining the load scenario corresponding to the demand response day that has occurred includes: obtaining a first time curve and a second time curve, the first time curve being a time curve formed by historical load data of each time period before and after the demand response on the demand response day that has occurred when demand response is performed on the demand response day that has occurred, and the second time curve being a time curve formed by historical load data before and after the demand response on the demand response day that has occurred when demand response is not performed on the demand response day that has occurred; for a first partial curve corresponding to the adjacent day after the demand response day that has occurred on the first time curve, and a second partial curve corresponding to the adjacent day after the demand response day that has occurred on the second time curve, if the first partial curve and the second partial curve have the same changing trend, then the load that needs to be adjusted on the demand response day that has occurred does not include a transferable load, and the load scenario corresponding to the demand response day that has occurred is the first load scenario; if the first partial curve and the second partial curve have different changing trends, then the load that needs to be adjusted on the demand response day that has occurred includes a transferable load, and the load scenario corresponding to the demand response day that has occurred is the second load scenario.

[0094] In some embodiments, the terminal obtains a first time curve and a second time curve, intercepts a first part of the curve on the first time curve for the day adjacent to the demand response day that has occurred and a second part of the curve on the second time curve for the day adjacent to the demand response day that has occurred, and compares the first part of the curve with the second part of the curve. If the change trends are the same, the load that needs to be adjusted on the demand response day that has occurred does not include a transferable load, and the corresponding scenario is the first load scenario; if the change trends are different, the load that needs to be adjusted on the demand response day that has occurred includes a transferable load, and the corresponding load scenario is the second load scenario.

[0095] Furthermore, the terminal may also obtain the first time curve and the second time curve, intercept the first part of the curve of the adjacent time period after the demand response day that has occurred on the first time curve and the second part of the curve of the adjacent time period after the demand response day that has occurred on the second time curve, and compare the first part of the curve with the second part of the curve. If the change trends are the same, the load that needs to be adjusted on the demand response day that has occurred does not include the transferable load, and the corresponding scenario is the first load scenario; if the change trends are different, the load that needs to be adjusted on the demand response day that has occurred includes the transferable load, and the corresponding load scenario is the second load scenario.

[0096] In the above embodiment, the demand response day that has occurred is specifically divided into two scenarios by combining the characteristics of residential electricity consumption, so that the final prediction result is more in line with reality.

[0097] In some embodiments, the terminal will determine a first load scenario and a second load scenario based on whether the load that needs to be adjusted on the demand response day that has occurred includes a transferable load. The first load scenario corresponds to the load that needs to be adjusted on the demand response day that has occurred and does not include a transferable load. The second load scenario corresponds to the load that needs to be adjusted on the demand response day that has occurred and includes a transferable load.

[0098] Exemplarily, in the first scenario, the cosine similarity between the load characteristic vector corresponding to the demand response day that has occurred and the load characteristic vector corresponding to each typical day in the first typical day set is calculated, and the specific calculation formula is shown in (6); in the second scenario, the cosine similarity between the load characteristic vector corresponding to the demand response day that has occurred and the load characteristic vector corresponding to each typical day in the first typical day set is calculated, and the specific calculation formula is shown in (7).

[0099]

[0100]

[0101] Furthermore, if the calculated cosine similarity is less than d, it means that the demand response day that has occurred does not match the typical day, and if the cosine similarity is greater than d, it means that the demand response day that has occurred matches the typical day successfully. Wherein, d is a preset value, and the specific value can be set by the user according to the actual situation, and this embodiment does not limit this.

[0102] In the above embodiment, the typical days in the first typical day set are further screened by cosine similarity, and the successfully matched typical days are screened out as the basis for the final prediction, so that the final result is more accurate.

[0103] In some exemplary embodiments, the load baseline prediction method also includes: calculating the cluster center in the second typical day set, the cluster center being a typical day in the second typical day set; calculating a preset range of the load baseline based on the historical load data in the cluster center; determining whether the load baseline of each time period in the demand response day to be predicted is within the preset range, if it is within the preset range, the load baseline of the corresponding time period remains unchanged; if it is not within the preset range, the load data of the corresponding time period in the cluster center is used to replace the load baseline of the corresponding time period.

[0104] In some embodiments, after predicting the load baseline, the terminal will further revise the result of the predicted load baseline.

[0105] Specifically, the terminal will cluster the second typical day set to obtain the cluster center, which is a typical day in the second typical day set. Assuming that the cluster center is the kth typical day, the load data of the kth typical day is , calculate the maximum load and minimum load of all typical days in the second typical day concentration , set the preset threshold and , and calculate the preset range. The specific calculation formula is shown in (8).

[0106]

[0107] Determine whether the load baseline of each time period of the demand response day to be predicted is within the preset range. If it is within the preset range, the load baseline of the corresponding time period remains unchanged; if it is not within the preset range, the load data of the corresponding time period in the cluster center is used to replace the load baseline of the corresponding time period.

[0108] In the above embodiment, the final load baseline result is corrected to prevent the prediction result from being overfitted or having a large prediction error, so that the final prediction result is more accurate.

[0109] To explain the load baseline prediction method in the present application in detail, an embodiment is used for explanation below. The specific flow chart is shown in FIG2 . By way of example, the present application is described with respect to the load baseline prediction method in a specific residential electricity usage scenario.

[0110] The terminal will obtain the historical load data of the demand response day and the demand response day that have occurred through the residential meter expansion module, and extract at least one of the maximum load, minimum load, average load, peak-to-valley difference rate or average change rate in the historical load data. The specific calculation formula for extracting the historical load characteristics of the demand response day from the historical load data is shown in (3), and the specific calculation formula for extracting the historical load characteristics of the day after the demand response is executed on the demand response day from the historical load data is shown in (4). Based on the historical load characteristics, the load characteristic vector corresponding to the demand response day that has occurred is obtained, and the specific calculation formula is shown in formula (5).

[0111] The meter expansion module pushes the load characteristic vector corresponding to the demand response day that has occurred to the relevant business system of the load aggregator or the power grid, determines the season of the demand response day that has occurred, and filters out the typical days in the same season as the demand response day that has occurred to obtain the first data set; determines the classification type of the demand response day that has occurred, if it is a working day, then filters the typical days that are working days in the first data set as the second data set; if the demand response day that has occurred is a non-working day, then filters the typical days that are non-working days in the first data set as the second data set.

[0112] Obtain the electricity consumption data of each typical day in the second data set and the electricity consumption data of the demand response day that has occurred when the demand response is not executed, and filter out the typical days in the second data set whose difference between the electricity consumption data of each typical day in the second data set and the electricity consumption data of the demand response day that has occurred when the demand response is not executed is less than a preset range, as the first typical day set. The preset range is determined by the terminal according to actual conditions, and this embodiment does not limit this. The number of typical days in the first typical day set is not less than n.

[0113] A first time curve and a second time curve are obtained, and a specific image is shown in FIG3 . The first time curve is a time curve formed by historical load data around the demand response day when demand response is performed on the demand response day that has occurred, and the second time curve is a time curve formed by historical load data around the demand response day when demand response is not performed on the demand response day that has occurred; for the first part of the curve corresponding to the adjacent day after the demand response day that has occurred on the first time curve, and the second part of the curve corresponding to the adjacent day after the demand response day that has occurred on the second time curve, if the first part of the curve has the same changing trend as the second part of the curve, then the load to be adjusted on the demand response day that has occurred does not include transferable load, and the load scenario corresponding to the demand response day that has occurred is the first load scenario; if the first part of the curve has different changing trends from the second part of the curve, then the load to be adjusted on the demand response day that has occurred includes transferable load, and the load scenario corresponding to the demand response day that has occurred is the second load scenario.

[0114] In the first scenario, the cosine similarity between the load characteristic vector corresponding to the demand response day that has occurred and the load characteristic vector corresponding to each typical day in the first typical day set is calculated, and the specific calculation formula is shown in (6); in the second scenario, the cosine similarity between the load characteristic vector corresponding to the demand response day that has occurred and the load characteristic vector corresponding to each typical day in the first typical day set is calculated, and the specific calculation formula is shown in (7). If the cosine similarity is less than d, it means that the demand response day that has occurred does not match the typical day, and if the cosine similarity is greater than d, it means that the demand response day that has occurred matches the typical day successfully. Further, the second typical day set is selected.

[0115] The prediction algorithm is used to predict the load baseline after the demand response day to be predicted based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set. The number of typical days in the second typical day set is not less than I. The prediction algorithm formula is shown in (1).

[0116] Cluster the second typical day set to obtain the cluster center, which is a typical day in the second typical day set. Assuming that the cluster center is the kth typical day, the load data of the kth typical day is , calculate the maximum load and minimum load of all typical days in the second typical day concentration , set the preset threshold and , and calculate the preset range. The specific calculation formula is shown in (8).

[0117] Determine whether the load baseline of each period of the demand response day to be predicted is within the preset range. If it is within the preset range, the load baseline of the corresponding period remains unchanged; if it is not within the preset range, the load data of the corresponding period in the cluster center is used to replace the load baseline of the corresponding period. After the correction is completed, the final resident load baseline prediction result is obtained.

[0118] This application considers the load characteristics of the non-demand response period on the demand response day, takes the characteristics as the basis for typical day matching, further combines the characteristics of residential electricity consumption, selects different load characteristics according to different scenarios, and screens the typical day loads that can be matched. Finally, the load data of the cluster center is used to replace the baseline load that is not accurately predicted, and the baseline load is corrected to make the prediction method more suitable for the specific scenario and the prediction result more accurate.

[0119] 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.

[0120] Based on the same inventive concept, the embodiment of the present application also provides a load baseline prediction device for implementing the load baseline prediction method involved above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more load baseline prediction device embodiments provided below can refer to the limitations of the load baseline prediction method above, and will not be repeated here.

[0121] In an exemplary embodiment, as shown in FIG4 , a load baseline prediction device is provided, including: an acquisition module 401 , a screening module 402 and a prediction module 403 , wherein:

[0122] An acquisition module is used to acquire historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data;

[0123] The acquisition module is further used to acquire a load characteristic vector corresponding to a demand response day that has occurred according to the historical load characteristics;

[0124] A screening module, used for screening typical days meeting preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set;

[0125] The screening module is further used to match the load characteristic vector corresponding to the demand response day that has occurred with the load characteristic vector corresponding to each typical day in the first typical day set based on different scenarios, and screen out the typical days in the first typical day set that are successfully matched as the second typical day set;

[0126] The prediction module is used to make a prediction based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

[0127] In some embodiments, the acquisition module is further used to acquire load characteristics, including at least one of the maximum load, minimum load, average load, peak-to-valley difference rate or average change rate in the load data.

[0128] In some embodiments, the above-mentioned screening module is also used to screen typical days that are in the same season as the demand response day that has occurred, to obtain a first data set; to screen out typical days with the same classification type as the demand response day that has occurred from the first data set, to obtain a second data set, wherein the classification types include working days and non-working days; and to screen out typical days whose similarity between the historical load data of the demand response day that has occurred and the historical load data reaches a preset degree from the second data set, to obtain a first typical day set.

[0129] In some embodiments, the device also includes a matching module for determining a load scenario corresponding to the demand response day that has occurred according to whether the load that needs to be adjusted on the demand response day that has occurred includes a transferable load; obtaining a load characteristic vector corresponding to the demand response day that has occurred according to the load scenario corresponding to the demand response day that has occurred; calculating the cosine similarity between the load characteristic vector corresponding to the demand response day that has occurred and the load characteristic vector corresponding to each typical day in the first typical day set; and taking a typical day with a cosine similarity greater than a preset value as a typical day that has been successfully matched in the first typical day set.

[0130] In some embodiments, the device also includes a determination module, which is used to obtain a first time curve and a second time curve, the first time curve being a time curve formed by historical load data around the demand response day that has occurred when demand response is performed on the demand response day that has occurred, and the second time curve being a time curve formed by historical load data around the demand response day that has occurred when demand response is not performed on the demand response day that has occurred; for the first partial curve corresponding to the adjacent day after the demand response day that has occurred on the first time curve, and the second partial curve corresponding to the adjacent day after the demand response day that has occurred on the second time curve, if the first partial curve and the second partial curve have the same changing trend, then the load that needs to be adjusted on the demand response day that has occurred does not include transferable load, and the load scenario corresponding to the demand response day that has occurred is the first load scenario; if the first partial curve and the second partial curve have different changing trends, then the load that needs to be adjusted on the demand response day that has occurred includes transferable load, and the load scenario corresponding to the demand response day that has occurred is the second load scenario.

[0131] In some embodiments, the device also includes a correction module, which is used to calculate the cluster center in the second typical day set, where the cluster center is a typical day in the second typical day set; calculate the preset range of the load baseline based on the historical load data in the cluster center; determine whether the load baseline of each time period in the demand response day to be predicted is within the preset range. If it is within the preset range, the load baseline of the corresponding time period remains unchanged; if it is not within the preset range, the load data of the corresponding time period in the cluster center is used to replace the load baseline of the corresponding time period.

[0132] Each module in the above load baseline prediction device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0133] In an exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be shown in FIG5. 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 via a system bus, and the communication interface, the display unit, and the input device are connected to the system bus via 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 may 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 baseline prediction method is implemented. The display unit of the computer device is used to form a visually visible picture, which may 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.

[0134] Those skilled in the art will understand that the structure shown in FIG. 5 is merely a block diagram of a partial 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 shown in the figure, or combine certain components, or have a different arrangement of components.

[0135] 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 following steps are implemented:

[0136] Obtain historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data;

[0137] According to the historical load characteristics, a load characteristic vector corresponding to the demand response day that has occurred is obtained;

[0138] Selecting typical days that meet preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set;

[0139] Based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and the typical days that are successfully matched in the first typical day set are selected as the second typical day set;

[0140] Prediction is performed based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

[0141] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0142] Obtain historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data;

[0143] According to the historical load characteristics, a load characteristic vector corresponding to the demand response day that has occurred is obtained;

[0144] Selecting typical days that meet preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set;

[0145] Based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and the typical days that are successfully matched in the first typical day set are selected as the second typical day set;

[0146] Prediction is performed based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

[0147] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0148] Obtain historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data;

[0149] According to the historical load characteristics, a load characteristic vector corresponding to the demand response day that has occurred is obtained;

[0150] Selecting typical days that meet preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set;

[0151] Based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and the typical days that are successfully matched in the first typical day set are selected as the second typical day set;

[0152] Prediction is performed based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

[0153] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0154] 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.

[0155] 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.

[0156] 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 load baseline prediction method, characterized in that: The method comprises: Obtain historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data; According to the historical load characteristics, a load characteristic vector corresponding to the demand response day that has occurred is obtained; Selecting typical days that meet preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set; Based on different scenarios, the load characteristic vector corresponding to the demand response day that has occurred is matched with the load characteristic vector corresponding to each typical day in the first typical day set, and the typical days that are successfully matched in the first typical day set are selected as the second typical day set; Prediction is performed based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

2. The method according to claim 1, characterized in that The load characteristic includes at least one of the maximum load, the minimum load, the average load, the peak-to-valley difference rate or the average change rate in the load data.

3. The method according to claim 1, characterized in that The step of selecting typical days meeting preset conditions as the first typical day set includes: Screening typical days that are in the same season as the demand response days that have occurred to obtain a first data set; Screening out typical days with the same classification type as the demand response day that has occurred from the first data set to obtain a second data set, wherein the classification type includes working days and non-working days; From the second data set, typical days whose similarity with the historical load data of the demand response day that has occurred reaches a preset degree are screened out to obtain a first typical day set.

4. The method according to claim 1, characterized in that: The matching of the load characteristic vector corresponding to the demand response day that has occurred with the load characteristic vector corresponding to each typical day in the first typical day set based on different scenarios, and screening out typical days in the first typical day set that are successfully matched, includes: Determine the load scenario corresponding to the demand response day that has occurred based on whether the load that needs to be adjusted on the demand response day that has occurred includes transferable load; According to the load scenario corresponding to the demand response day that has occurred, a load characteristic vector corresponding to the demand response day that has occurred is obtained; Calculate the cosine similarity between the load characteristic vector corresponding to the demand response day that has occurred and the load characteristic vector corresponding to each typical day in the first typical day set; The typical day whose cosine similarity is greater than a preset value is taken as the typical day successfully matched in the first typical day set.

5. The method according to claim 4, characterized in that The load scenario corresponding to the demand response day that has occurred is determined according to whether the load to be adjusted on the demand response day that has occurred includes the transferable load, including: Obtaining a first time curve and a second time curve, wherein the first time curve is a time curve formed by historical load data around the demand response day that has occurred, in the case where demand response is performed on the demand response day that has occurred, and the second time curve is a time curve formed by historical load data around the demand response day that has occurred, in the case where demand response is not performed on the demand response day that has occurred; For the first partial curve corresponding to the first time curve on the adjacent day after the demand response day that has occurred, and the second partial curve corresponding to the second time curve on the adjacent day after the demand response day that has occurred, if the change trends of the first partial curve and the second partial curve are the same, the load that needs to be adjusted on the demand response day that has occurred does not include the transferable load, and the load scenario corresponding to the demand response day that has occurred is the first load scenario; if the change trends of the first partial curve and the second partial curve are different, the load that needs to be adjusted on the demand response day that has occurred includes the transferable load, and the load scenario corresponding to the demand response day that has occurred is the second load scenario.

6. The method according to claim 1, characterized in that The method further comprises: Calculating a cluster center in the second typical day set, where the cluster center is a typical day in the second typical day set; Calculating a preset range of a load baseline based on historical load data in the cluster center; Determine whether the load baseline of each time period within the demand response day to be predicted is within a preset range. If it is within the preset range, the load baseline of the corresponding time period remains unchanged; if it is not within the preset range, the load data of the corresponding time period in the cluster center is used to replace the load baseline of the corresponding time period.

7. A load baseline prediction device, characterized in that: The device comprises: An acquisition module is used to acquire historical load data before and on the demand response day that has occurred, and extract historical load characteristics before and on the demand response day based on the historical load data; The acquisition module is further used to acquire a load characteristic vector corresponding to a demand response day that has occurred according to the historical load characteristics; A screening module, used for screening typical days meeting preset conditions as a first typical day set, and extracting a load characteristic vector corresponding to each typical day in the first typical day set; The screening module is further used to match the load characteristic vector corresponding to the demand response day that has occurred with the load characteristic vector corresponding to each typical day in the first typical day set based on different scenarios, and screen out the typical days in the first typical day set that are successfully matched as the second typical day set; The prediction module is used to make a prediction based on the historical load data before the demand response day to be predicted and the historical load data of each typical day in the second typical day set to obtain the load baseline of the demand response day to be predicted.

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