Temperature-sensitive load prediction method, device, equipment, and storage medium

Abstract

The present application discloses a temperature-sensitive load forecasting method, apparatus, device, and storage medium. The method includes: obtaining a first power consumption set and a first temperature data set for a target area in a historical time period; determining a baseline power consumption based on the first power consumption set and the first temperature data set; obtaining a second power consumption set and a second temperature data set for the target area in a current time period; determining a first temperature-sensitive power consumption set based on the second power consumption set and the baseline power consumption; determining a third temperature data set for the target area in a future time period; inputting the second temperature data set, the first temperature-sensitive power consumption set, and the third temperature data set into a temperature-sensitive power consumption forecasting model to obtain a target temperature-sensitive power consumption set for the future time period; and determining d target temperature-sensitive loads corresponding to d future sub-time periods based on the third temperature data set and the target temperature-sensitive power consumption set. The present application is conducive to improving the accuracy of load forecasting.

Description

Technical Field

[0001] The present application relates to the technical field of load forecasting, and in particular to a method, apparatus, device, and storage medium for temperature-sensitive load forecasting. Background Art

[0002] In modern urban distribution networks, balancing supply and demand is crucial for ensuring the safe and stable operation of power systems. With global climate change and accelerating urbanization, the impact of high summer temperatures on electricity demand is becoming increasingly significant. In particular, temperature-sensitive loads continue to increase, primarily due to the widespread use of air conditioning and refrigeration equipment. This significant increase in temperature-sensitive loads often leads to greater power supply pressure on the grid during peak load periods, resulting in power shortages and reduced power quality.

[0003] Load forecasting is the main means to solve the above problems. Traditional load forecasting methods usually predict load data for future time periods by analyzing historical load data. However, they fail to fully consider the direct impact of high temperature weather on temperature-sensitive electricity consumption and temperature-sensitive loads, resulting in low load forecasting accuracy. Summary of the Invention

[0004] In order to solve the above-mentioned problems existing in the prior art, the embodiments of the present application provide a temperature-sensitive load prediction method, apparatus, device and storage medium. The benchmark power consumption is determined by the first power consumption set and the first temperature data set of the target area in the historical time period, and the first temperature-sensitive power consumption set of the target area in the current time period is determined according to the second power consumption set and the benchmark power consumption of the current time period. Then, the target temperature-sensitive power consumption set of the target area in the future time period is predicted by the second temperature data set of the current time period, the first temperature-sensitive power consumption set and the third temperature data set of the future time period, and d target temperature-sensitive loads corresponding to d sub-future time periods of the target area in the future time period are determined. The predicted target temperature-sensitive loads take into account the impact of temperature on load, thereby improving the accuracy of load prediction.

[0005] In a first aspect, an embodiment of the present application provides a temperature-sensitive load prediction method, comprising:

[0006] Obtain a first electricity consumption set and a first temperature dataset for a target area during a historical time period; the historical time period includes a number of working days; the first electricity consumption set includes a number of first electricity consumptions, with each of the a number of working days corresponding one to the first electricity consumptions; the first temperature dataset includes t temperatures corresponding to the a number of working days, with each working day corresponding to at least one temperature; a is a positive integer, and t is an integer greater than a;

[0007] determining a baseline power consumption based on the first power consumption set and the first temperature data set;

[0008] Obtain a second power consumption set and a second temperature dataset for the target area in a current time period; the current time period includes b working days before the current working day; the second power consumption set includes b second power consumptions, and the b working days have a one-to-one correspondence with the b second power consumptions; the second temperature dataset includes h temperatures corresponding to the b working days, with each working day corresponding to at least one temperature; b is a positive integer, and h is an integer greater than b;

[0009] determining, based on the second power consumption set and the benchmark power consumption, a first temperature-sensitive power consumption set for the target area during the current time period; the first temperature-sensitive power consumption set comprising b first temperature-sensitive power consumptions, the b working days corresponding one-to-one to the b first temperature-sensitive power consumptions;

[0010] Determining a third temperature dataset for the target area in a future time period; the future time period includes c working days after the current working day; the third temperature dataset includes y temperatures corresponding to the c working days, with each working day corresponding to at least one temperature; c is a positive integer, and y is an integer greater than c;

[0011] Inputting the second temperature dataset, the first temperature-sensitive power consumption set, and the third temperature dataset into a temperature-sensitive power consumption prediction model to obtain a target temperature-sensitive power consumption set for the target area in the future time period; the target temperature-sensitive power consumption set includes c target temperature-sensitive power consumptions, and the c working days correspond one-to-one to the c target temperature-sensitive power consumptions;

[0012] Based on the third temperature data set and the target temperature-sensitive power consumption set, determine d target temperature-sensitive loads corresponding to d sub-future time periods of the target area in the future time period; each working day in the future time period includes at least one sub-future time period, each sub-future time period corresponds to a target temperature-sensitive load, and d is an integer greater than c.

[0013] In a second aspect, an embodiment of the present application provides a temperature-sensitive load prediction device, comprising:

[0014] an acquisition unit, configured to acquire a first electricity consumption set and a first temperature data set for a target area during a historical time period; the historical time period includes a number of working days; the first electricity consumption set includes a number of first electricity consumptions, with the a number of working days corresponding one to one with the a number of first electricity consumptions; the first temperature data set includes t number of temperatures corresponding to the a number of working days, with each working day corresponding to at least one temperature; a is a positive integer, and t is an integer greater than a;

[0015] a processing unit, configured to determine a reference power consumption according to the first power consumption set and the first temperature data set;

[0016] The acquisition unit is configured to acquire a second power consumption set and a second temperature data set for the target area in a current time period; the current time period includes b working days before the current working day; the second power consumption set includes b second power consumptions, and the b working days correspond one-to-one to the b second power consumptions; the second temperature data set includes h temperatures corresponding to the b working days, with each working day corresponding to at least one temperature; b is a positive integer, and h is an integer greater than b;

[0017] the processing unit being configured to determine, based on the second power consumption set and the benchmark power consumption, a first temperature-sensitive power consumption set for the target area during the current time period; the first temperature-sensitive power consumption set comprising b first temperature-sensitive power consumptions, the b working days corresponding one-to-one to the b first temperature-sensitive power consumptions;

[0018] Determining a third temperature dataset for the target area in a future time period; the future time period includes c working days after the current working day; the third temperature dataset includes y temperatures corresponding to the c working days, with each working day corresponding to at least one temperature; c is a positive integer, and y is an integer greater than c;

[0019] Inputting the second temperature dataset, the first temperature-sensitive power consumption set, and the third temperature dataset into a temperature-sensitive power consumption prediction model to obtain a target temperature-sensitive power consumption set for the target area in the future time period; the target temperature-sensitive power consumption set includes c target temperature-sensitive power consumptions, and the c working days correspond one-to-one to the c target temperature-sensitive power consumptions;

[0020] Based on the third temperature data set and the target temperature-sensitive power consumption set, determine d target temperature-sensitive loads corresponding to d sub-future time periods of the target area in the future time period; each working day in the future time period includes at least one sub-future time period, each sub-future time period corresponds to a target temperature-sensitive load, and d is an integer greater than c.

[0021] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor and a memory, wherein the processor is connected to the memory, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the electronic device performs the method described in the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method as described in the first aspect.

[0023] In a fifth aspect, an embodiment of the present application provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program product is operable to enable a computer to execute the method described in the first aspect.

[0024] The implementation of the embodiments of the present application has the following beneficial effects:

[0025] In an embodiment of the present application, first, a first electricity consumption set and a first temperature data set of the target area in a historical time period are obtained. Based on the first electricity consumption set and the first temperature data set, a baseline electricity consumption can be determined. Then, a second electricity consumption set and a second temperature data set of the target area in the current time period are obtained. Based on the second electricity consumption set and the baseline electricity consumption, a first temperature-sensitive electricity consumption set of the target area in the current time period can be determined. Furthermore, a third temperature data set of the target area in the future time period is determined, and the second temperature data set, the first temperature-sensitive electricity consumption set, and the third temperature data set are input into the temperature-sensitive electricity consumption prediction model to obtain a target temperature-sensitive electricity consumption set of the target area in the future time period. Finally, based on the third temperature data set and the target temperature-sensitive electricity consumption set, d target temperature-sensitive loads corresponding to d sub-future time periods of the target area in the future time period can be determined. Therefore, the benchmark power consumption can be determined through the power consumption and temperature data of the target area in the historical time period, and the temperature-sensitive power consumption of the current time period can be determined based on the power consumption and the benchmark power consumption of the current time period. Therefore, the temperature-sensitive power consumption of the future time period can be predicted based on the temperature data and temperature-sensitive power consumption of the current time period, as well as the temperature data of the future time period, and then the temperature-sensitive load of the future time period can be determined. The predicted temperature-sensitive load takes into account the impact of temperature data on power consumption, thereby improving the accuracy of load prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0027] Figure 1 A schematic diagram of an application scenario of a temperature-sensitive load prediction method provided in an embodiment of the present application;

[0028] Figure 2 A flow chart of a temperature-sensitive load prediction method provided in an embodiment of the present application;

[0029] Figure 3 A schematic diagram of a temperature-sensitive load prediction device provided in an embodiment of the present application;

[0030] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0032] The terms "first," "second," "third," and "fourth," etc., in the specification and claims of this application and the accompanying drawings are used to distinguish between different objects, rather than to describe a specific order. In addition, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0033] References herein to "embodiments" mean that a particular feature, result, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0034] First, see Figure 1 , Figure 1 Schematic diagram of an application scenario for a temperature-sensitive load forecasting method provided in an embodiment of the present application. The distribution network is a power network directly connected to the power demand side of the power system and constructed by overhead lines, cables, distribution transformers, reactive power compensation devices, and other related facilities. It is primarily used to distribute electrical energy in the transmission system to various power demand sides through distribution equipment.

[0035] like Figure 1As shown, the distribution network is directly connected to the power demand side in the target area, which can include residential and business users. The electrical devices on the power demand side receive and consume electricity from the distribution network to ensure normal operation of the electrical devices. These devices can include air conditioners, fans, water heaters, elevators, and so on. The total power consumed by all electrical devices in the target area is the target area's load. The total amount of energy consumed by all electrical devices in the target area during a time period is the target area's electricity consumption during that time period.

[0036] In an embodiment of the present application, smart meters are provided on the distribution lines of the power distribution network and the power demand side to collect the power consumption corresponding to the power demand side. The smart meters on the distribution lines can be communicatively connected to the processing device. Among them, the processing device can be a server for high-performance data processing and caching functions, including: application servers, database servers, virtual private servers (VPS), cloud servers, etc., which are not limited in this application. The processing device can also be an integrated circuit, chip or processor for data processing, data transmission and reception, and data storage functions, including: a central processing unit (CPU), a digital signal processor (DSP), a microcontroller unit (MCU), etc., which are not limited in this application. The processing device can also be an electronic device integrated with a processor, a transceiver and a memory, such as a smart phone, a computer, etc., which are not limited in this application.

[0037] The processing device can obtain real-time electricity consumption at the target area's power demand side from smart meters connected to the distribution network and the target area's distribution lines, and store the target area's electricity consumption and the time of collection in a memory. When performing load forecasting, the processing device can retrieve the target area's electricity consumption during a historical time period from the memory and, based on this historical electricity consumption, predict the target area's electricity consumption for a future time period. This can then be used to determine the target area's load for the future time period based on the target area's electricity consumption during the future time period.

[0038] However, electricity consumption on the power demand side can vary significantly under different temperature conditions. For example, during high summer temperatures, electricity consumption in the target area will increase significantly due to the heavy use of high-probability electrical equipment such as air conditioners and refrigeration equipment on the power demand side. The portion of electricity consumption that increases significantly with rising temperatures is called temperature-sensitive electricity consumption, and the load corresponding to temperature-sensitive electricity consumption is called temperature-sensitive load. Therefore, load forecasting based solely on electricity consumption over historical time periods is inaccurate. To prevent power shortages or reduced power quality in the distribution network due to increased temperature-sensitive electricity consumption, it is necessary to forecast temperature-sensitive electricity consumption and temperature-sensitive loads.

[0039] To this end, applied to the above scenario, in the temperature-sensitive load forecasting method provided in the present application, the processing device obtains a first power consumption set and a first temperature data set for a target area in a historical time period;

[0040] The processing device determines a reference power consumption according to the first power consumption set and the first temperature data set;

[0041] The processing device obtains a second power consumption set and a second temperature data set of the target area in the current time period;

[0042] The processing device determines a first temperature-sensitive power consumption set for the target area in the current time period based on the second power consumption set and the benchmark power consumption;

[0043] The processing device determines a third temperature data set for the target area in a future time period;

[0044] The processing device inputs the second temperature data set, the first temperature-sensitive power consumption set, and the third temperature data set into the temperature-sensitive power consumption prediction model to obtain a target temperature-sensitive power consumption set for the target area in the future time period;

[0045] The processing device determines, based on the third temperature data set and the target temperature-sensitive power consumption set, d target temperature-sensitive loads corresponding to d sub-future time periods in the target area in the future time period; each working day in the future time period includes at least one sub-future time period, each sub-future time period corresponds to a target temperature-sensitive load, and d is an integer greater than c.

[0046] It can be seen that, when applied to the above scenario, the processing device can determine the baseline power consumption based on the first power consumption set and the first temperature data set of the target area in the historical time period, and determine the first temperature-sensitive power consumption set of the target area in the current time period based on the second power consumption set and the baseline power consumption of the current time period. Then, based on the second temperature data set and the first temperature-sensitive power consumption set of the target area in the current time period, as well as the third temperature data set of the target area in the future time period, the target temperature-sensitive power consumption set of the target area in the future time period is predicted, thereby determining the target temperature-sensitive load of the target area in the future time period. The predicted target temperature-sensitive load fully considers the impact of temperature on load, thereby improving the accuracy of load prediction.

[0047] See Figure 2 , Figure 2 This is a flow chart of a temperature-sensitive load prediction method provided in an embodiment of the present application. The method is applied to the processing equipment in the above scenario, and the method includes but is not limited to the following steps:

[0048] 201: Obtain a first power consumption set and a first temperature data set for a target area in a historical time period.

[0049] In this embodiment of the present application, the historical time period includes a number of working days. The first power consumption set includes a number of first power consumptions, with each working day corresponding to each number of first power consumptions. The first temperature data set includes t temperatures corresponding to each working day, with each working day corresponding to at least one temperature. a is a positive integer, and t is an integer greater than a.

[0050] It should be noted that the first electricity consumption set and the first temperature data set for the historical time period are primarily used to determine the target area's baseline electricity consumption for each weekday. Therefore, the historical time period should be a period of mild temperatures where temperature has little or no impact on electricity consumption. For example, the historical time period could include all weekdays in April. During holidays, electricity consumption can be supplied normally due to residents being away from home and businesses suspending production, eliminating any supply shortages. Therefore, only weekday electricity consumption and temperature data are considered when forecasting temperature-sensitive loads.

[0051] The processing device retrieves from the memory the first power consumption corresponding to each working day in the historical time period in the target area, and at least one temperature corresponding to each working day in the historical time period, to obtain a first power consumption set and a first temperature data set. The at least one temperature corresponding to each working day is at least one temperature in at least one historical time period corresponding to the working day, for example, the temperature from 10:00 to 11:00 and the temperature from 11:00 to 12:00 on a working day.

[0052] 202: Determine a baseline power consumption according to the first power consumption set and the first temperature data set.

[0053] In the embodiment of the present application, the baseline power consumption represents the average power consumption of the target area on each working day when the temperature is less than a first preset temperature. The first preset temperature is the highest temperature at which the power consumption of the target area is not affected by temperature. The first preset temperature can be pre-analyzed based on historical data.

[0054] Exemplarily, determining the baseline power consumption according to the first power consumption set and the first temperature data set may include:

[0055] Obtain the maximum temperature corresponding to each of the a working days from the t temperatures in the first temperature dataset;

[0056] Determine a working day in which the maximum temperature is less than or equal to the first preset temperature as the first working day, and obtain m first working days;

[0057] Determine the power consumption corresponding to each of the m first working days from the first power consumption set to obtain m third power consumptions;

[0058] The reference power consumption is determined according to the m third power consumptions.

[0059] In the embodiment of the present application, m is a positive integer less than or equal to a. The processing device can determine the baseline power consumption of the target area by using the power consumption on working days with a maximum temperature less than or equal to the first preset temperature.

[0060] Specifically, the processing device first obtains the maximum temperature corresponding to each working day from the t temperatures corresponding to the a working days in the first temperature dataset. For example, if the temperatures corresponding to working day d1 include: 23, 25, 26, 28, 25, and 19, the processing device obtains the maximum temperature for working day d1 as 28. In this way, the processing device can obtain the maximum temperature corresponding to each of the a working days.

[0061] Then, the processing device obtains working days with a maximum temperature less than or equal to the first preset temperature from the a working days, obtaining m first working days. The power consumption corresponding to each working day in the m first working days is obtained from the first power consumption set, obtaining m third power consumptions.

[0062] Finally, the processing device determines the average power consumption of the m third power consumptions to obtain the benchmark power consumption, that is, the average power consumption of the target area on each working day under the condition that it is not affected by temperature.

[0063] It can be seen that the processing equipment can select m first working days with the highest temperature less than or equal to the first preset temperature from a working days, and determine the baseline power consumption of the target area based on the m third power consumption corresponding to the m first working days. Based on the baseline power consumption, the power consumption of the target area that increases under the influence of temperature, that is, the temperature-sensitive power consumption, can be determined. Therefore, through the correspondence between the temperature-sensitive power consumption and temperature, the temperature-sensitive power consumption of the future time period can be predicted, thereby improving the accuracy of the temperature-sensitive power consumption prediction.

[0064] 203: Obtain a second power consumption set and a second temperature data set for the target area in the current time period.

[0065] In this embodiment of the present application, the current time period includes b working days before the current working day. The second power consumption set includes b second power consumptions, with b working days corresponding to b second power consumptions. The second temperature data set includes h temperatures corresponding to b working days, with each working day corresponding to at least one temperature. b is a positive integer, and h is an integer greater than b. Optionally, b can be 10.

[0066] It should be noted that in order to improve the prediction accuracy of temperature-sensitive electricity consumption, the processing device needs to obtain the second electricity consumption set and the second temperature data set of the target area b working days before the current working day, so as to predict the electricity consumption of the target area in the future time period based on the electricity consumption changes and temperature changes in the b working days before the current working day.

[0067] 204: Determine a first temperature-sensitive power consumption set of the target area in the current time period based on the second power consumption set and the benchmark power consumption.

[0068] In this embodiment of the present application, the first temperature-sensitive power usage set includes b first temperature-sensitive power usages, and b working days correspond one-to-one to b first temperature-sensitive power usages. The first temperature-sensitive power usage is the difference between the second power usage corresponding to the first temperature-sensitive power usage and the baseline power usage. The processing device can determine the first temperature-sensitive power usage set based on the b second power usages in the second power usage set and the baseline power usage.

[0069] Exemplarily, determining the first temperature-sensitive power consumption set of the target area in the current time period based on the second power consumption set and the benchmark power consumption may include:

[0070] Determine the difference between each second power consumption in the second power consumption set and the benchmark power consumption as the third temperature-sensitive power consumption, and obtain b third temperature-sensitive power consumptions;

[0071] Adjusting the third temperature-sensitive power consumptions that are less than the preset temperature-sensitive power consumption among the b third temperature-sensitive power consumptions to the preset temperature-sensitive power consumption to obtain b fourth temperature-sensitive power consumptions;

[0072] Obtain the maximum temperature corresponding to each of the b working days from the h temperatures in the second temperature dataset;

[0073] Determine the working day with the highest temperature less than or equal to the second preset temperature among the b working days as the second working day, and adjust the fourth temperature-sensitive power consumption corresponding to the second working day among the b fourth temperature-sensitive power consumptions to the preset temperature-sensitive power consumption to obtain a first temperature-sensitive power consumption set.

[0074] In the embodiment of the present application, b working days correspond one-to-one to b third temperature-sensitive power consumptions, and b working days correspond one-to-one to b fourth temperature-sensitive power consumptions.

[0075] Specifically, the processing device first determines the difference between each of the b second power consumptions in the second power consumption set and the reference power consumption to obtain b third temperature-sensitive power consumptions.

[0076] It should be noted that the practical engineering significance of temperature-sensitive electricity consumption refers to the increased electricity consumption caused by a significant increase in temperature. Temperature-sensitive electricity consumption should increase with increasing temperature and decrease with decreasing temperature. Therefore, temperature-sensitive electricity consumption will not have a negative value.

[0077] Based on this, the processing device adjusts the b third temperature-sensitive power consumptions that are less than the preset temperature-sensitive power consumption to the preset temperature-sensitive power consumption, thereby obtaining b fourth temperature-sensitive power consumptions. Optionally, the preset temperature-sensitive power consumption may be 0. The preset temperature-sensitive power consumption may be 0 or may be pre-set based on actual weather conditions in the target area.

[0078] It should be noted that when the maximum temperature on a working day is less than the preset temperature, the temperature-sensitive power consumption will also be reduced to 0. Therefore, the processing device also needs to adjust the fourth temperature-sensitive power consumption corresponding to the working day with the maximum temperature less than or equal to the second preset temperature to the preset temperature-sensitive power consumption. Specifically, the processing device first obtains the maximum temperature corresponding to each working day from the h temperatures corresponding to the b working days in the second temperature data set. Then, all second working days with the maximum temperature less than or equal to the second preset temperature are selected from the b working days, and the fourth temperature-sensitive power consumption corresponding to the second working day among the b fourth temperature-sensitive power consumptions is adjusted to the preset temperature-sensitive power consumption, obtaining b first temperature-sensitive power consumptions, that is, obtaining a first temperature-sensitive power consumption set. Among them, the second preset temperature can be equal to the first preset temperature, and optionally, the second preset temperature can be 0.

[0079] As can be seen, by determining the difference between each second power consumption in the second power consumption set and the baseline power consumption, the processing device can obtain b third temperature-sensitive power consumptions. By adjusting the temperature-sensitive power consumption of the b third temperature-sensitive power consumptions that is less than the preset temperature-sensitive power consumption, b fourth temperature-sensitive power consumptions can be obtained. By adjusting the fourth temperature-sensitive power consumption corresponding to weekdays with a maximum temperature less than or equal to the second preset temperature, the first temperature-sensitive power consumption set can be obtained. Thus, by correcting the temperature-sensitive power consumption, a more accurate first temperature-sensitive power consumption set can be obtained, thereby improving the accuracy of temperature-sensitive power consumption and temperature-sensitive load forecasts.

[0080] 205: Determine a third temperature dataset of the target area in a future time period.

[0081] In this embodiment of the present application, the future time period includes c working days following the current working day. The third temperature dataset includes y temperatures corresponding to the c working days, with at least one temperature corresponding to each working day. c is a positive integer, and y is an integer greater than c. It should be understood that the temperature-sensitive power consumption forecast in this application is a short-term forecast, that is, the forecast starts from the working day after the current working day.

[0082] Optionally, the processing device can obtain a third temperature dataset for the target area in a future time period from predicted temperature data released by the meteorological department. It should be noted that since the temperatures in the third temperature dataset are all predicted temperatures, the prediction of temperature-sensitive power consumption based on the third temperature dataset inherently contains errors. Ensuring that the error is within a preset error range can help the distribution network allocate electricity appropriately and address the power supply shortage caused by the increase in temperature-sensitive loads.

[0083] 206: Input the second temperature data set, the first temperature-sensitive power consumption set, and the third temperature data set into a temperature-sensitive power consumption prediction model to obtain a target temperature-sensitive power consumption set for the target area in a future time period.

[0084] In this embodiment of the present application, the target temperature-sensitive power consumption set includes c target temperature-sensitive power consumptions, and c working days correspond one-to-one to c target temperature-sensitive power consumptions. The temperature-sensitive power consumption prediction model is a model built based on a random forest algorithm. The processing device inputs the second temperature dataset, the first temperature-sensitive power consumption set, and the third temperature dataset into the temperature-sensitive power consumption prediction model, and uses the temperature-sensitive power consumption prediction model to predict the target temperature-sensitive power consumption set for the target area in the future time period.

[0085] Exemplarily, before inputting the second temperature data set, the first temperature-sensitive power consumption set, and the third temperature data set into the temperature-sensitive power consumption prediction model, the following steps may also be included:

[0086] Obtain historical sample sets;

[0087] Divide the historical sample set into training set and test set;

[0088] Randomly sample the data in the training set to obtain i sample sets;

[0089] Generate i decision trees based on i sample sets, with each sample set corresponding to one decision tree, specifically including: obtaining a target sample set and generating a reference decision tree based on the target sample set; determining a target feature set corresponding to the target sample set; randomly selecting j features from the target feature set, and using the set consisting of j features as a feature subset of the reference decision tree; selecting partitioning features from the feature subset according to a preset partitioning criterion, and partitioning the reference decision tree according to the partitioning features to obtain a target decision tree;

[0090] According to i decision trees, determine the initial temperature-sensitive electricity consumption prediction model;

[0091] The initial temperature-sensitive electricity consumption prediction model is tested on the test set to obtain the prediction error;

[0092] A temperature-sensitive electricity consumption prediction model is determined based on the prediction error and the initial temperature-sensitive electricity consumption prediction model.

[0093] In an embodiment of the present application, the historical sample set includes multiple fourth temperature data sets and multiple second temperature-sensitive power consumption sets corresponding to multiple first sample time periods, as well as multiple fifth temperature data sets and multiple third temperature-sensitive power consumption sets corresponding to multiple second sample time periods. The multiple first sample time periods correspond one-to-one with the multiple second sample time periods. Each first sample time period includes b working days prior to the sample working day corresponding to the first sample time period, and each second sample time period includes c working days after the sample working day of the first sample time period corresponding to the second sample time period. The fourth temperature data set includes h temperatures corresponding to the b working days corresponding to the fourth temperature data set, with each working day corresponding to at least one temperature. The second temperature-sensitive power consumption set includes b second temperature-sensitive power consumption corresponding to the b working days corresponding to the second temperature-sensitive power consumption set, with the b working days corresponding one-to-one with the b second temperature-sensitive power consumption. The fifth temperature data set includes y temperatures corresponding to the c working days corresponding to the fifth temperature data set, with each working day corresponding to at least one temperature. The third temperature-sensitive power consumption set includes c target temperature-sensitive power consumptions corresponding to the third temperature-sensitive power consumption set, and the c target temperature-sensitive power consumptions correspond one-to-one to the c working days corresponding to the third temperature-sensitive power consumption set.

[0094] Where i and j are both positive integers. The target sample set is any sample set among the i sample sets. The target feature set includes all features corresponding to the target sample set. The target decision tree is the decision tree corresponding to the target sample set among the i decision trees.

[0095] Specifically, the processing device first needs to obtain a historical sample set, where the historical sample set can be sample data from the q years preceding the current forecast year, where q is equal to the number of first sample time periods and equal to the number of second sample time periods. The multiple first sample time periods are the q time periods corresponding to the current time period in the q years preceding it, and the multiple second sample time periods are the q time periods corresponding to the future time period in the q years preceding it. The processing device divides the historical sample set into a training set and a test set. For example, the data corresponding to the year preceding the current forecast year in the historical sample set is used as the training set, and the remaining data is used as the test set.

[0096] The processing device then randomly samples the data in the training set. The number of data samples each time can be pre-set. For example, 5 sample data are sampled each time, and the dimension of each sample data can be 21. By randomly sampling the data in the training set, i sample sets can be obtained. i is the total number of samplings; optionally, i can be 100. It should be noted that since the data in the i sample sets are all randomly sampled, the sample sets are diverse. Constructing a temperature-sensitive electricity consumption prediction model based on the i sample sets can reduce the risk of overfitting of the temperature-sensitive electricity consumption prediction model and improve the accuracy of temperature-sensitive electricity consumption prediction.

[0097] Furthermore, the processing device generates i decision trees based on the i sample sets. Each decision tree generated can be used to predict temperature-sensitive electricity consumption. Finally, the prediction results of the i decision trees are combined to output the final predicted temperature-sensitive electricity consumption. Taking the generation of a target decision tree based on a target sample set as an example, the processing device obtains the target sample set and generates a reference decision tree based on the target sample set. The root node of the reference decision tree is used to store the target sample set. Then, a target feature set consisting of all features corresponding to the target sample set is determined, wherein the type of feature extracted from each sample data in the target sample set can be pre-set. Furthermore, the processing device randomly selects j features from the target feature set and uses the set consisting of j features as the feature subset of the reference decision tree. j can be pre-set based on the actual prediction accuracy requirements. Optionally, j can be 122. Randomly selecting features can make the selected features diverse, thereby reducing the risk of overfitting of the temperature-sensitive electricity consumption prediction model and improving the accuracy of temperature-sensitive electricity consumption prediction. Next, the processing device selects partitioning features from the feature subset according to the preset partitioning criteria, and partitions the nodes of the reference decision tree according to the partitioning features to generate new leaf nodes on the reference decision tree, which are used to store the partitioning features. Among them, the preset partitioning criteria may include, for example, any of the following partitioning criteria: feature partitioning based on information gain, feature partitioning based on the Gini index, feature partitioning based on variance reduction, and so on. In this way, by regressively partitioning the reference decision tree features until the decision tree reaches the preset depth, a target decision tree corresponding to the target sample set can be obtained. Based on the above-mentioned method for generating the target decision tree, i decision trees can be generated.

[0098] The processing device then integrates the i decision trees constructed and connects their outputs to a combiner to generate an initial temperature-sensitive power consumption prediction model. In this initial temperature-sensitive power consumption prediction model, each of the i decision trees can predict temperature-sensitive power consumption, resulting in a predicted temperature-sensitive power consumption corresponding to that decision tree. The combiner can determine the final output of temperature-sensitive power consumption based on the predicted temperature-sensitive power consumption corresponding to the i decision trees. For example, the average of the i predicted temperature-sensitive power consumption can be used as the final output of temperature-sensitive power consumption.

[0099] After obtaining the initial temperature-sensitive power consumption prediction model, the processing device can test the initial temperature-sensitive power consumption prediction model through the test set, and determine the prediction error based on the temperature-sensitive power consumption predicted by the initial temperature-sensitive power consumption prediction model and the actual temperature-sensitive power consumption in the test set. Optionally, the processing device can determine the mean absolute percentage error (MAPE) between the predicted temperature-sensitive power consumption and the actual temperature-sensitive power consumption, and use the mean absolute percentage error as the prediction error. When the prediction error is less than or equal to the preset error, the initial temperature-sensitive power consumption prediction model is used as the temperature-sensitive power consumption prediction model. When the prediction error is greater than the preset error, the initial temperature-sensitive power consumption prediction model is continued to be trained until the prediction error is less than or equal to the preset error, and the temperature-sensitive power consumption prediction model is obtained.

[0100] As can be seen, by dividing the historical sample set into a training set and a test set and randomly sampling the training set to obtain i sample sets, the sample set can be diversified, reducing the risk of overfitting the temperature-sensitive electricity consumption prediction model and improving prediction accuracy. Then, through random sampling and feature partitioning, i decision trees can be generated from the i sample sets. This can diversify the feature subsets, reduce the risk of overfitting the temperature-sensitive electricity consumption prediction model, and improve prediction accuracy. Based on the i decision trees, an initial temperature-sensitive electricity consumption prediction model can be determined. Furthermore, the initial temperature-sensitive electricity consumption prediction model can be tested on the test set to obtain the prediction error. Finally, based on the prediction error and the initial temperature-sensitive electricity consumption prediction model, a temperature-sensitive electricity consumption prediction model can be obtained. Using this temperature-sensitive electricity consumption prediction model to predict temperature-sensitive electricity consumption can improve the accuracy of temperature-sensitive electricity consumption prediction.

[0101] After obtaining the temperature-sensitive power consumption prediction model, the processing device may input the second temperature dataset, the first temperature-sensitive power consumption set, and the third temperature dataset into the temperature-sensitive power consumption prediction model to obtain a target temperature-sensitive power consumption set for the target area in the future time period. It is understood that the temperature-sensitive power consumption prediction model can only predict one target temperature-sensitive power consumption at a time. Therefore, the temperature-sensitive power consumption prediction model needs to perform a cycle of prediction c times based on the time sequence of c working days to obtain c target temperature-sensitive power consumptions, i.e., the target temperature-sensitive power consumption set.

[0102] 207 : Determine d target temperature-sensitive loads corresponding to d sub-future time periods in the target area in the future time period according to the third temperature data set and the target temperature-sensitive power consumption set.

[0103] In this embodiment of the present application, each working day in the future time period includes at least one sub-future time period, each sub-future time period corresponds to a target temperature-sensitive load, and d is an integer greater than c. It will be appreciated that the above-mentioned predicted target temperature-sensitive power consumption set is the temperature-sensitive power consumption corresponding to each of the c working days, and the processing device also needs to determine the temperature-sensitive power consumption and temperature-sensitive load for different sub-future time periods under each working day in order to formulate a power allocation plan corresponding to each sub-future time period.

[0104] In an optional embodiment, before determining, based on the third temperature data set and the target temperature-sensitive power consumption set, d target temperature-sensitive loads corresponding to d sub-future time periods of the target area in the future time period, the following method may also be included:

[0105] determining, based on the third temperature data set, an average temperature for each of the c working days;

[0106] If there are k consecutive working days among the c working days, and the corresponding k average temperatures are all greater than the third preset temperature, then based on the k average temperatures, the average temperature difference between any two adjacent working days among the k working days is determined to obtain the k-1 average temperature difference;

[0107] Determine k-1 proportional coefficients based on k-1 average temperature differences;

[0108] Based on k-1 proportional coefficients, determine the k-1 temperature-sensitive electricity consumption growth;

[0109] According to the k-1 temperature-sensitive power consumption growth, the k-1 target temperature-sensitive power consumption corresponding to the k-1 working day in the k working days is adjusted to obtain the adjusted k-1 target temperature-sensitive power consumption.

[0110] In this embodiment, each working day corresponds to an average temperature. k is an integer greater than x and less than or equal to c, where x is a positive integer less than c. x represents the minimum number of working days required for continuous high temperatures to cause a change in the target temperature-sensitive power consumption in the target area. This value can be pre-set based on actual test results.

[0111] Specifically, the processing device first obtains all temperatures corresponding to each of the c working days from the third temperature data set, and then determines the average temperature of all temperatures of each working day.

[0112] If, among c working days, there are k consecutive working days with k average temperatures corresponding to them exceeding the third preset temperature—for example, if the average temperature of four consecutive working days, Monday, Tuesday, Wednesday, and Thursday, is greater than the third preset temperature—the processing device will sequentially determine the difference between any two adjacent average temperatures among the k average temperatures, based on the chronological order of the k working days. This difference is the average temperature difference between any two adjacent working days among the k working days, thereby obtaining the k-1 average temperature difference. The third preset temperature represents the highest average temperature that causes a change in electricity consumption in the target area and can be pre-set based on actual test results.

[0113] It is understandable that when k average temperatures for k consecutive working days are all greater than the third preset temperature, the impact of temperature on the power consumption of the target area will continue to increase.

[0114] Therefore, the processing device determines k-1 proportionality coefficients based on the k-1 average temperature differences, one proportionality coefficient for each average temperature difference. These proportionality coefficients represent the proportionality of the increase in temperature-sensitive power consumption with temperature. The mapping between average temperature differences and proportionality coefficients can be pre-set based on actual test results. The processing device calculates the k-1 temperature-sensitive power consumption increase by multiplying each of the k-1 proportionality coefficients by the target temperature-sensitive power consumption for the workday corresponding to that proportionality coefficient.

[0115] Finally, the processing device sums each temperature-sensitive power consumption increase in the k-1 temperature-sensitive power consumption increase with the target temperature-sensitive power consumption of the working day corresponding to the temperature-sensitive power consumption increase, obtains the adjusted k-1 target temperature-sensitive power consumption, and uses the adjusted k-1 target temperature-sensitive power consumption as the k-1 target temperature-sensitive power consumption.

[0116] Thus, by determining that k average temperatures corresponding to k consecutive working days among c working days are all greater than the third preset temperature, the processing device determines k-1 proportional coefficients based on the average temperature difference between any two adjacent working days among the k working days. Based on the k-1 proportional coefficients, the processing device determines k-1 temperature-sensitive power consumption increases to adjust the k-1 target temperature-sensitive power consumption, thereby obtaining the adjusted k-1 target temperature-sensitive power consumption. The adjusted target temperature-sensitive power consumption takes into account the impact of consecutive high temperatures on the temperature-sensitive power consumption, thereby making the adjusted target temperature-sensitive power consumption closer to the actual temperature-sensitive power consumption and improving the accuracy of the temperature-sensitive power consumption prediction.

[0117] Exemplarily, determining, based on the third temperature data set and the target temperature-sensitive power consumption set, d target temperature-sensitive loads corresponding to d sub-future time periods of the target area in the future time period may include:

[0118] Obtaining d temperatures greater than a first preset temperature from the y temperatures in the third temperature data set;

[0119] Obtain d sub-future time periods corresponding to d temperatures from the future time period;

[0120] Determine the difference between each of the d temperatures and the first preset temperature to obtain d temperature differences;

[0121] Based on the d temperature differences, determine the total temperature difference corresponding to each of the c working days, and obtain c total temperature differences;

[0122] Determine d second temperature-sensitive power consumptions based on the d temperature differences, the c total temperature differences, and the target temperature-sensitive power consumption set;

[0123] According to the d second temperature-sensitive power consumptions and the d future sub-time periods, d target temperature-sensitive loads are determined.

[0124] In this embodiment of the present application, each of the c working days corresponds to at least one of the d temperatures. Each of the c working days corresponds to at least one of the d temperature differences. The d sub-future time periods correspond one-to-one with the d second temperature-sensitive power consumptions.

[0125] Specifically, the processing device first obtains d temperatures greater than the first preset temperature from the y temperatures in the third temperature data set. The d temperatures represent the temperatures corresponding to different time periods on each of the c working days. Based on the d temperatures, the processing device can divide the future time period into d sub-future time periods corresponding to the d temperatures. Each sub-future time period is a time period on the working day corresponding to the sub-future time period. For example, on a working day, the temperature from 10 a.m. to 11 a.m. and the temperature from 11 a.m. to 12 a.m. are both greater than the first preset temperature. The processing device will treat 10 a.m. to 11 a.m. as a sub-future time period and 11 a.m. to 12 a.m. as a sub-future time period.

[0126] Then, the processing device determines the difference between each of the d temperatures and the first preset temperature, obtaining d temperature differences. Based on all the temperature differences corresponding to each of the c working days, the total temperature difference corresponding to each of the c working days is determined, obtaining c total temperature differences. The processing device uses the ratio of each of the d temperature differences to the total temperature difference corresponding to the temperature difference as the distribution density, obtaining d distribution densities. Each of the c working days corresponds to at least one of the d distribution densities. Next, the processing device uses the product of the d distribution densities and the target temperature-sensitive power consumption of the working day corresponding to the distribution density as the second temperature-sensitive power consumption, obtaining d second temperature-sensitive power consumptions.

[0127] Finally, the processing device obtains the duration of each of the d future sub-time periods. The ratio of each of the d second temperature-sensitive power consumptions to the duration of the sub-time period corresponding to the second temperature-sensitive power consumption is used as the temperature-sensitive load for the future sub-time period, thereby obtaining d target temperature-sensitive loads.

[0128] Therefore, by adjusting the temperature-sensitive power consumption of the sub-future time periods in the future time period whose temperature is greater than the first preset temperature, the target temperature-sensitive power consumption of each sub-future time period is obtained, and the target temperature-sensitive load is determined, thereby fully considering the impact of temperature on temperature-sensitive power consumption and temperature-sensitive load, and improving the accuracy of temperature-sensitive load prediction.

[0129] Exemplarily, the method may further include:

[0130] Obtaining a residential electricity consumption set and an enterprise electricity consumption set of a target area in a historical time period according to the first electricity consumption set;

[0131] Determine f historical residential electricity consumption time periods corresponding to f residential electricity consumption, and determine g historical enterprise electricity consumption time periods corresponding to g enterprise electricity consumption;

[0132] Obtain d historical residential electricity consumption time periods corresponding to d sub-future time periods from f historical residential electricity consumption time periods, and obtain d residential electricity consumption corresponding to d historical residential electricity consumption time periods from f residential electricity consumption;

[0133] Obtain d historical enterprise electricity consumption time periods corresponding to d future sub-time periods from g historical enterprise electricity consumption time periods, and obtain d enterprise electricity consumption corresponding to d historical enterprise electricity consumption time periods from g enterprise electricity consumption;

[0134] According to the electricity consumption of d residents and d enterprises, determine the electricity consumption weights of d residents and d enterprises;

[0135] According to d residential electricity consumption weights, d enterprise electricity consumption weights, and d target temperature-sensitive loads, d residential temperature-sensitive loads and d enterprise temperature-sensitive loads corresponding to d sub-future time periods are determined.

[0136] In this embodiment of the present application, the residential electricity consumption set includes f residential electricity consumptions and g business electricity consumptions corresponding to a business days. Each business day corresponds to at least one residential electricity consumption and at least one business electricity consumption. f and g are positive integers greater than a. Each residential electricity consumption corresponds to a historical residential electricity consumption time period, and each business electricity consumption corresponds to a historical business electricity consumption time period.

[0137] It should be noted that the electricity demand side in the target area includes residential electricity consumption and corporate electricity consumption, among which residential electricity consumption and corporate electricity consumption correspond to different electricity consumption time periods. Therefore, the processing equipment will determine d residential temperature-sensitive loads and d corporate temperature-sensitive loads corresponding to d sub-future time periods.

[0138] Specifically, the processing device first obtains a residential electricity consumption set and a business electricity consumption set for the target area during a historical time period based on the first electricity consumption set. The residential electricity consumption set is a set of electricity consumption collected by smart meters on residential power lines, and the business electricity consumption set is a set of electricity consumption collected by smart meters on business power lines. Both the residential electricity consumption set and the business electricity consumption set are stored in a memory. The processing device can obtain the residential electricity consumption set and the business electricity consumption set for the historical time period based on the first electricity consumption set.

[0139] Then, f historical residential electricity consumption time periods corresponding to the f residential electricity consumptions in the residential electricity consumption set are determined, and g historical enterprise electricity consumption time periods corresponding to the g enterprise electricity consumptions in the enterprise electricity consumption set are determined. Furthermore, d historical residential electricity consumption time periods corresponding to the d future sub-time periods are obtained from the f historical residential electricity consumption time periods, and d residential electricity consumptions corresponding to the d historical residential electricity consumption time periods are obtained from the f residential electricity consumptions. d historical enterprise electricity consumption time periods corresponding to the d future sub-time periods are obtained from the g historical enterprise electricity consumption time periods, and d enterprise electricity consumptions corresponding to the d historical enterprise electricity consumption time periods are obtained from the g enterprise electricity consumptions.

[0140] Furthermore, the processing device determines the sum of residential electricity consumption and enterprise electricity consumption corresponding to each of the d future sub-time periods as the total electricity consumption corresponding to the sub-time period. The ratio of the residential electricity consumption corresponding to each of the d future sub-time periods to the total electricity consumption corresponding to the sub-time period is used as the residential electricity consumption weight corresponding to the residential electricity consumption, thereby obtaining d residential electricity consumption weights. The ratio of the enterprise electricity consumption corresponding to each of the d future sub-time periods to the total electricity consumption corresponding to the sub-time period is used as the enterprise electricity consumption weight corresponding to the enterprise electricity consumption, thereby obtaining d enterprise electricity consumption weights.

[0141] Finally, the processing device determines the product of each residential electricity weight among the d residential electricity weights and the target temperature-sensitive load for the sub-future time period corresponding to the residential electricity weight as the residential temperature-sensitive load for the sub-future time period, thereby obtaining d residential temperature-sensitive loads. The processing device determines the product of each enterprise electricity weight among the d enterprise electricity weights and the target temperature-sensitive load for the sub-future time period corresponding to the enterprise electricity weight as the enterprise temperature-sensitive load for the sub-future time period, thereby obtaining d enterprise temperature-sensitive loads.

[0142] Therefore, the residential electricity consumption weight and the enterprise electricity consumption weight are determined through the residential electricity consumption and the enterprise electricity consumption in the historical time period, and the residential temperature-sensitive load and the enterprise temperature-sensitive load are determined based on the residential electricity consumption weight and the enterprise electricity consumption weight. The changes of the residential temperature-sensitive load and the enterprise temperature-sensitive load in the future time period can be predicted, so that the power system can better distribute electricity.

[0143] In summary, in an embodiment of the present application, the first electricity consumption set and the first temperature data set of the target area in the historical time period are first obtained. Based on the first electricity consumption set and the first temperature data set, the baseline electricity consumption can be determined. Then, the second electricity consumption set and the second temperature data set of the target area in the current time period are obtained. Based on the second electricity consumption set and the baseline electricity consumption, the first temperature-sensitive electricity consumption set of the target area in the current time period can be determined. Furthermore, the third temperature data set of the target area in the future time period is determined, and the second temperature data set, the first temperature-sensitive electricity consumption set and the third temperature data set are input into the temperature-sensitive electricity consumption prediction model to obtain the target temperature-sensitive electricity consumption set of the target area in the future time period. Finally, based on the third temperature data set and the target temperature-sensitive electricity consumption set, d target temperature-sensitive loads corresponding to d sub-future time periods of the target area in the future time period can be determined. Therefore, the benchmark power consumption can be determined through the power consumption and temperature data of the target area in the historical time period, and the temperature-sensitive power consumption of the current time period can be determined based on the power consumption and the benchmark power consumption of the current time period. Therefore, the temperature-sensitive power consumption of the future time period can be predicted based on the temperature data and temperature-sensitive power consumption of the current time period, as well as the temperature data of the future time period, and then the temperature-sensitive load of the future time period can be determined. The predicted temperature-sensitive load takes into account the impact of temperature data on power consumption, thereby improving the accuracy of load prediction.

[0144] See Figure 3 , Figure 3 Schematic diagram of a temperature-sensitive load prediction device provided in an embodiment of the present application. The temperature-sensitive load prediction device 300 can be the processing device of any of the above embodiments. The temperature-sensitive load prediction device 300 includes an acquisition unit 301 and a processing unit 302.

[0145] An acquisition unit 301 is configured to acquire a first power consumption set and a first temperature data set for a target area during a historical time period; the historical time period includes a number of working days; the first power consumption set includes a number of first power consumptions, with a working day corresponding to a number of first power consumptions; the first temperature data set includes t temperatures corresponding to a number of working days, with each working day corresponding to at least one temperature; a is a positive integer, and t is an integer greater than a.

[0146] The processing unit 302 is configured to determine a reference power consumption according to the first power consumption set and the first temperature data set;

[0147] An acquisition unit 301 is configured to acquire a second power consumption set and a second temperature data set for a target area in a current time period; the current time period includes b working days before the current working day; the second power consumption set includes b second power consumptions, with b working days corresponding to the b second power consumptions in a one-to-one relationship; the second temperature data set includes h temperatures corresponding to the b working days, with each working day corresponding to at least one temperature; b is a positive integer, and h is an integer greater than b;

[0148] The processing unit 302 is configured to determine a first temperature-sensitive power consumption set for the target area in the current time period based on the second power consumption set and the benchmark power consumption; the first temperature-sensitive power consumption set includes b first temperature-sensitive power consumptions, and b working days correspond to b first temperature-sensitive power consumptions in a one-to-one manner;

[0149] Determine a third temperature dataset for the target area in a future time period; the future time period includes c working days after the current working day; the third temperature dataset includes y temperatures corresponding to the c working days, with each working day corresponding to at least one temperature; c is a positive integer, and y is an integer greater than c;

[0150] Inputting the second temperature data set, the first temperature-sensitive power consumption set, and the third temperature data set into the temperature-sensitive power consumption prediction model to obtain a target temperature-sensitive power consumption set for the target area in the future time period; the target temperature-sensitive power consumption set includes c target temperature-sensitive power consumptions, and c working days correspond to c target temperature-sensitive power consumptions in a one-to-one manner;

[0151] Based on the third temperature data set and the target temperature-sensitive power consumption set, determine d target temperature-sensitive loads corresponding to d sub-future time periods in the target area in the future time period; each working day in the future time period includes at least one sub-future time period, each sub-future time period corresponds to a target temperature-sensitive load, and d is an integer greater than c.

[0152] In a possible embodiment, in determining, based on the third temperature data set and the target temperature-sensitive power consumption set, d target temperature-sensitive loads corresponding to d sub-future time periods in the target area in the future time period, the processing unit 302 is specifically configured to:

[0153] Obtaining d temperatures greater than a first preset temperature from the y temperatures in the third temperature data set; each of the c working days corresponds to at least one temperature in the d temperatures;

[0154] Obtain d sub-future time periods corresponding to d temperatures from the future time period;

[0155] Determine the difference between each of the d temperatures and the first preset temperature to obtain d temperature differences; each of the c working days corresponds to at least one temperature difference in the d temperature differences;

[0156] Based on the d temperature differences, determine the total temperature difference corresponding to each of the c working days, and obtain c total temperature differences;

[0157] Determine d second temperature-sensitive power consumptions based on the d temperature differences, the c total temperature differences, and the target temperature-sensitive power consumption set; the d sub-future time periods correspond one-to-one to the d second temperature-sensitive power consumptions;

[0158] According to the d second temperature-sensitive power consumptions and the d future sub-time periods, d target temperature-sensitive loads are determined.

[0159] In a possible embodiment, in determining the reference power consumption according to the first power consumption set and the first temperature data set, the processing unit 302 is specifically configured to:

[0160] Obtain the maximum temperature corresponding to each of the a working days from the t temperatures in the first temperature dataset;

[0161] Determine a working day with a maximum temperature less than or equal to a first preset temperature as the first working day, and obtain m first working days, where m is a positive integer less than or equal to a;

[0162] Determine the power consumption corresponding to each of the m first working days from the first power consumption set to obtain m third power consumptions;

[0163] The reference power consumption is determined according to the m third power consumptions.

[0164] In a possible embodiment, in determining the first temperature-sensitive power consumption set of the target area in the current time period based on the second power consumption set and the benchmark power consumption, the processing unit 302 is specifically configured to:

[0165] Determine the difference between each second power consumption and the benchmark power consumption in the second power consumption set as the third temperature-sensitive power consumption, and obtain b third temperature-sensitive power consumptions; b working days correspond one to one with the b third temperature-sensitive power consumptions;

[0166] Adjusting the third temperature-sensitive power consumptions that are less than the preset temperature-sensitive power consumption among the b third temperature-sensitive power consumptions to the preset temperature-sensitive power consumption to obtain b fourth temperature-sensitive power consumptions; the b working days correspond one to one with the b fourth temperature-sensitive power consumptions;

[0167] Obtain the maximum temperature corresponding to each of the b working days from the h temperatures in the second temperature dataset;

[0168] Determine the working day with the highest temperature less than or equal to the second preset temperature among the b working days as the second working day, and adjust the fourth temperature-sensitive power consumption corresponding to the second working day among the b fourth temperature-sensitive power consumptions to the preset temperature-sensitive power consumption to obtain a first temperature-sensitive power consumption set.

[0169] In a possible embodiment, before inputting the second temperature dataset, the first temperature-sensitive power consumption set, and the third temperature dataset into the temperature-sensitive power consumption prediction model, the processing unit 302 is further configured to:

[0170] Obtain a historical sample set; the historical sample set includes a plurality of fourth temperature data sets and a plurality of second temperature-sensitive power consumption sets corresponding to a plurality of first sample time periods, and a plurality of fifth temperature data sets and a plurality of third temperature-sensitive power consumption sets corresponding to a plurality of second sample time periods; the plurality of first sample time periods correspond one-to-one to the plurality of second sample time periods; each first sample time period includes b working days before a sample working day corresponding to the first sample time period, and each second sample time period includes c working days after a sample working day of the first sample time period corresponding to the second sample time period;

[0171] Divide the historical sample set into training set and test set;

[0172] Randomly sample the data in the training set to obtain i sample sets; i is a positive integer;

[0173] Generate i decision trees based on i sample sets, where each sample set corresponds to a decision tree, specifically including: obtaining a target sample set, and generating a reference decision tree based on the target sample set, where the target sample set is any sample set in the i sample sets; determining a target feature set corresponding to the target sample set, where the target feature set includes all features corresponding to the target sample set; randomly selecting j features from the target feature set, and using the set consisting of j features as a feature subset of the reference decision tree, where j is a positive integer; selecting partitioning features from the feature subset according to a preset partitioning criterion, and partitioning the reference decision tree according to the partitioning features to obtain a target decision tree, where the target decision tree is the decision tree corresponding to the target sample set in the i decision trees;

[0174] According to i decision trees, determine the initial temperature-sensitive electricity consumption prediction model;

[0175] The initial temperature-sensitive electricity consumption prediction model is tested on the test set to obtain the prediction error;

[0176] A temperature-sensitive electricity consumption prediction model is determined based on the prediction error and the initial temperature-sensitive electricity consumption prediction model.

[0177] In a possible embodiment, before determining, based on the third temperature data set and the target temperature-sensitive power consumption set, d target temperature-sensitive loads corresponding to d sub-future time periods in the target area in the future time period, the processing unit 302 is further configured to:

[0178] determining, based on the third temperature data set, an average temperature for each of the c working days;

[0179] If there are k consecutive working days among c working days where the k average temperatures corresponding to them are all greater than the third preset temperature, then based on the k average temperatures, determine the average temperature difference between any two adjacent working days among the k working days to obtain the k-1 average temperature difference; each working day corresponds to an average temperature; k is an integer greater than x and less than or equal to c, and x is a positive integer less than c;

[0180] Determine k-1 proportional coefficients based on k-1 average temperature differences;

[0181] Based on k-1 proportional coefficients, determine the k-1 temperature-sensitive electricity consumption growth;

[0182] According to the k-1 temperature-sensitive power consumption growth, the k-1 target temperature-sensitive power consumption corresponding to the k-1 working day in the k working days is adjusted to obtain the adjusted k-1 target temperature-sensitive power consumption.

[0183] In a possible embodiment, the processing unit 302 is further configured to:

[0184] Obtain a residential electricity consumption set and an enterprise electricity consumption set for the target area during a historical time period based on the first electricity consumption set; the residential electricity consumption set includes f residential electricity consumptions and g enterprise electricity consumptions corresponding to a working days; each working day corresponds to at least one residential electricity consumption and at least one enterprise electricity consumption; f and g are positive integers greater than a;

[0185] Determine f historical residential electricity consumption time periods corresponding to f residential electricity consumptions, and determine g historical enterprise electricity consumption time periods corresponding to g enterprise electricity consumptions; each residential electricity consumption corresponds to a historical residential electricity consumption time period, and each enterprise electricity consumption corresponds to a historical enterprise electricity consumption time period;

[0186] Obtain d historical residential electricity consumption time periods corresponding to d sub-future time periods from f historical residential electricity consumption time periods, and obtain d residential electricity consumption corresponding to d historical residential electricity consumption time periods from f residential electricity consumption;

[0187] Obtain d historical enterprise electricity consumption time periods corresponding to d future sub-time periods from g historical enterprise electricity consumption time periods, and obtain d enterprise electricity consumption corresponding to d historical enterprise electricity consumption time periods from g enterprise electricity consumption;

[0188] According to the electricity consumption of d residents and d enterprises, determine the electricity consumption weights of d residents and d enterprises;

[0189] According to d residential electricity consumption weights, d enterprise electricity consumption weights, and d target temperature-sensitive loads, d residential temperature-sensitive loads and d enterprise temperature-sensitive loads corresponding to d sub-future time periods are determined.

[0190] See Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 4 As shown, electronic device 400 includes a transceiver 401, a processor 402, and a memory 403. These are connected via a bus 404. Memory 403 is used to store computer programs and data and transmit data stored in memory 403 to processor 402. Electronic device 400 may be temperature-sensitive load forecasting device 300. Electronic device 400 may also be the processing device of any of the above embodiments.

[0191] The processor 402 is configured to read the computer program in the memory 403 and perform the following operations:

[0192] Obtaining a first power consumption set and a first temperature data set for a target area in a historical time period;

[0193] Determining a baseline power consumption based on the first power consumption set and the first temperature data set;

[0194] Obtain a second power consumption set and a second temperature data set for the target area in the current time period;

[0195] Determining a first temperature-sensitive power consumption set for the target area in the current time period based on the second power consumption set and the benchmark power consumption;

[0196] determining a third temperature dataset for the target area at a future time period;

[0197] Inputting the second temperature data set, the first temperature-sensitive power consumption set, and the third temperature data set into the temperature-sensitive power consumption prediction model to obtain a target temperature-sensitive power consumption set for the target area in the future time period;

[0198] According to the third temperature data set and the target temperature-sensitive power consumption set, d target temperature-sensitive loads corresponding to d sub-future time periods in the target area in the future time period are determined.

[0199] The above mainly introduces the solution of the embodiment of the present application from the perspective of the execution process of the method side. It is understandable that, in order to realize the above functions, the electronic device includes a hardware structure and / or software module corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiment provided herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software driven hardware manner depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0200] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement part or all of the steps of any one of the methods described in the above method embodiments.

[0201] An embodiment of the present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute part or all of the steps of any one of the methods described in the above method embodiments.

[0202] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all optional embodiments, and the actions and modules involved are not necessarily required for this application.

[0203] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0204] In the several embodiments provided in this application, it should be understood that the disclosed devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0205] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0206] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or in the form of software program modules.

[0207] If the integrated unit is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a memory and includes a number of instructions for enabling a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned memory includes various media that can store program codes, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0208] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program. The program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0209] The above is a detailed introduction to the embodiments of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for those skilled in the art, according to the idea of ​​the present application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A temperature-sensitive load forecasting method, characterized in that: include: Obtain a first electricity consumption set and a first temperature dataset for a target area during a historical time period; the historical time period includes a number of working days; the first electricity consumption set includes a number of first electricity consumptions, with the a number of working days corresponding one to one with the a number of first electricity consumptions; the first temperature dataset includes t number of temperatures corresponding to the a number of working days, with each working day corresponding to at least one temperature; a is a positive integer, and t is an integer greater than a; determining a baseline power consumption according to the first power consumption set and the first temperature data set; Obtain a second power consumption set and a second temperature dataset for the target area in a current time period; the current time period includes b working days before the current working day; the second power consumption set includes b second power consumptions, and the b working days have a one-to-one correspondence with the b second power consumptions; the second temperature dataset includes h temperatures corresponding to the b working days, with each working day corresponding to at least one temperature; b is a positive integer, and h is an integer greater than b; Determining a first temperature-sensitive power consumption set for the target area during the current time period based on the second power consumption set and the benchmark power consumption; The first temperature-sensitive power consumption set includes b first temperature-sensitive power consumptions, and the b working days correspond one-to-one to the b first temperature-sensitive power consumptions; Determining a third temperature dataset for the target area in a future time period; the future time period includes c working days after the current working day; the third temperature dataset includes y temperatures corresponding to the c working days, with each working day corresponding to at least one temperature; c is a positive integer, and y is an integer greater than c; Inputting the second temperature data set, the first temperature-sensitive power consumption set, and the third temperature data set into a temperature-sensitive power consumption prediction model to obtain a target temperature-sensitive power consumption set for the target area in the future time period; The target temperature-sensitive power consumption set includes c target temperature-sensitive power consumptions, and the c working days correspond one-to-one to the c target temperature-sensitive power consumptions; Based on the third temperature data set and the target temperature-sensitive power consumption set, determine d target temperature-sensitive loads corresponding to d sub-future time periods of the target area in the future time period; each working day in the future time period includes at least one sub-future time period, each sub-future time period corresponds to a target temperature-sensitive load, and d is an integer greater than c.

2. The method according to claim 1, characterized in that The determining, based on the third temperature data set and the target temperature-sensitive power consumption set, d target temperature-sensitive loads corresponding to d future sub-time periods of the target area in the future time period includes: Obtaining d temperatures greater than a first preset temperature from the y temperatures in the third temperature data set; each of the c working days corresponds to at least one of the d temperatures; Obtaining d sub-future time periods corresponding to the d temperatures from the future time period; Determine the difference between each of the d temperatures and the first preset temperature to obtain d temperature differences; each of the c working days corresponds to at least one of the d temperature differences; Determine the total temperature difference corresponding to each of the c working days based on the d temperature differences, to obtain c total temperature differences; determining d second temperature-sensitive power consumptions according to the d temperature differences, the c total temperature differences, and the target temperature-sensitive power consumption set; wherein the d sub-future time periods correspond one-to-one to the d second temperature-sensitive power consumptions; The d target temperature-sensitive loads are determined according to the d second temperature-sensitive power consumptions and the d future sub-time periods.

3. The method according to claim 2, characterized in that The determining, according to the first power consumption set and the first temperature data set, a reference power consumption, includes: Obtaining the maximum temperature corresponding to each of the a working days from the t temperatures in the first temperature dataset; Determine a working day in which the maximum temperature is less than or equal to the first preset temperature among the a working days as a first working day, obtaining m first working days, where m is a positive integer less than or equal to a; Determine the power consumption corresponding to each of the m first working days from the first power consumption set to obtain m third power consumptions; The reference power consumption is determined according to the m third power consumptions.

4. The method according to any one of claims 1 to 3, characterized in that The determining, based on the second power consumption set and the benchmark power consumption, a first temperature-sensitive power consumption set for the target area in the current time period includes: Determine that the difference between each second power consumption in the second power consumption set and the benchmark power consumption is a third temperature-sensitive power consumption, and obtain b third temperature-sensitive power consumptions; the b working days correspond one-to-one to the b third temperature-sensitive power consumptions; adjusting the third temperature-sensitive power consumptions that are less than the preset temperature-sensitive power consumption among the b third temperature-sensitive power consumptions to the preset temperature-sensitive power consumption, to obtain b fourth temperature-sensitive power consumptions; the b working days correspond one-to-one to the b fourth temperature-sensitive power consumptions; Obtaining the maximum temperature corresponding to each of the b working days from the h temperatures in the second temperature dataset; Determine the working day among the b working days whose maximum temperature is less than or equal to the second preset temperature as the second working day, and adjust the fourth temperature-sensitive power consumption corresponding to the second working day among the b fourth temperature-sensitive power consumptions to the preset temperature-sensitive power consumption, to obtain the first temperature-sensitive power consumption set.

5. The method according to any one of claims 1 to 3, characterized in that Before inputting the second temperature data set, the first temperature-sensitive power consumption set, and the third temperature data set into the temperature-sensitive power consumption prediction model, the method further includes: Obtain a historical sample set; the historical sample set includes a plurality of fourth temperature data sets and a plurality of second temperature-sensitive power consumption sets corresponding to a plurality of first sample time periods, and a plurality of fifth temperature data sets and a plurality of third temperature-sensitive power consumption sets corresponding to a plurality of second sample time periods; the plurality of first sample time periods correspond one-to-one to the plurality of second sample time periods; each first sample time period includes b working days before a sample working day corresponding to the first sample time period, and each second sample time period includes c working days after a sample working day of the first sample time period corresponding to the second sample time period; Dividing the historical sample set into a training set and a test set; Randomly sample the data in the training set to obtain i sample sets; i is a positive integer; Generate i decision trees based on the i sample sets, with each sample set corresponding to a decision tree, specifically comprising: obtaining a target sample set, and generating a reference decision tree based on the target sample set, where the target sample set is any one of the i sample sets; determining a target feature set corresponding to the target sample set, where the target feature set includes all features corresponding to the target sample set; randomly selecting j features from the target feature set, and using the set consisting of the j features as a feature subset of the reference decision tree, where j is a positive integer; selecting partitioning features from the feature subset according to a preset partitioning criterion, and partitioning the reference decision tree according to the partitioning features to obtain a target decision tree, where the target decision tree is the decision tree corresponding to the target sample set among the i decision trees; Determine an initial temperature-sensitive power consumption prediction model based on the i decision trees; Testing the initial temperature-sensitive power consumption prediction model using the test set to obtain a prediction error; The temperature-sensitive power consumption prediction model is determined according to the prediction error and the initial temperature-sensitive power consumption prediction model.

6. The method according to any one of claims 1 to 3, characterized in that Before determining, based on the third temperature data set and the target temperature-sensitive power consumption set, the d target temperature-sensitive loads corresponding to the d future sub-time periods of the target area in the future time period, the method further includes: determining, based on the third temperature data set, an average temperature of each of the c working days; If there are k consecutive working days among the c working days, and the k average temperatures corresponding to them are all greater than the third preset temperature, then, based on the k average temperatures, determine the average temperature difference between any two adjacent working days among the k working days to obtain a k-1 average temperature difference; each working day corresponds to an average temperature; k is an integer greater than x and less than or equal to c, and x is a positive integer less than c; Determining k-1 proportional coefficients according to the k-1 average temperature differences; Determining k-1 temperature-sensitive electricity consumption increases based on the k-1 proportional coefficients; According to the k-1 temperature-sensitive power consumption increase, the k-1 target temperature-sensitive power consumption corresponding to the k-1 working day in the k working days is adjusted to obtain the adjusted k-1 target temperature-sensitive power consumption.

7. The method according to any one of claims 1 to 3, characterized in that The method further comprises: Obtaining a residential electricity consumption set and a business electricity consumption set for the target area during the historical time period based on the first electricity consumption set; the residential electricity consumption set includes f residential electricity consumptions and g business electricity consumptions corresponding to the a working days; each working day corresponds to at least one residential electricity consumption and at least one business electricity consumption; f and g are positive integers greater than a; Determine f historical residential electricity consumption time periods corresponding to the f residential electricity consumptions, and determine g historical enterprise electricity consumption time periods corresponding to the g enterprise electricity consumptions; each residential electricity consumption corresponds to a historical residential electricity consumption time period, and each enterprise electricity consumption corresponds to a historical enterprise electricity consumption time period; Obtaining d historical residential electricity consumption time periods corresponding to the d future sub-time periods from the f historical residential electricity consumption time periods, and obtaining d residential electricity consumptions corresponding to the d historical residential electricity consumption time periods from the f residential electricity consumptions; Obtaining d historical enterprise electricity consumption time periods corresponding to the d future sub-time periods from the g historical enterprise electricity consumption time periods, and obtaining d enterprise electricity consumptions corresponding to the d historical enterprise electricity consumption time periods from the g enterprise electricity consumptions; Determining d residential electricity consumption weights and d enterprise electricity consumption weights based on the d residential electricity consumption and the d enterprise electricity consumption; The d residential temperature-sensitive loads and d enterprise temperature-sensitive loads corresponding to the d sub-future time periods are determined according to the d residential electricity consumption weights, the d enterprise electricity consumption weights, and the d target temperature-sensitive loads.

8. A temperature-sensitive load prediction device, characterized in that: include: an acquisition unit, configured to acquire a first electricity consumption set and a first temperature data set for a target area during a historical time period; the historical time period including a working days; the first electricity consumption set including a first electricity consumptions, the a working days corresponding one-to-one to the a first electricity consumptions; and the first temperature data set including t temperatures corresponding to the a working days, with each working day corresponding to at least one temperature; a is a positive integer, and t is an integer greater than a; a processing unit, configured to determine a reference power consumption according to the first power consumption set and the first temperature data set; The acquisition unit is configured to acquire a second power consumption set and a second temperature data set for the target area in a current time period; the current time period includes b working days before the current working day; the second power consumption set includes b second power consumptions, and the b working days correspond one-to-one to the b second power consumptions; the second temperature data set includes h temperatures corresponding to the b working days, with each working day corresponding to at least one temperature; b is a positive integer, and h is an integer greater than b; The processing unit is configured to determine a first temperature-sensitive power consumption set for the target area in the current time period based on the second power consumption set and the benchmark power consumption; The first temperature-sensitive power consumption set includes b first temperature-sensitive power consumptions, and the b working days correspond one-to-one to the b first temperature-sensitive power consumptions; Determining a third temperature dataset for the target area in a future time period; the future time period includes c working days after the current working day; the third temperature dataset includes y temperatures corresponding to the c working days, with each working day corresponding to at least one temperature; c is a positive integer, and y is an integer greater than c; Inputting the second temperature data set, the first temperature-sensitive power consumption set, and the third temperature data set into a temperature-sensitive power consumption prediction model to obtain a target temperature-sensitive power consumption set for the target area in the future time period; The target temperature-sensitive power consumption set includes c target temperature-sensitive power consumptions, and the c working days correspond one-to-one to the c target temperature-sensitive power consumptions; Based on the third temperature data set and the target temperature-sensitive power consumption set, determine d target temperature-sensitive loads corresponding to d sub-future time periods of the target area in the future time period; each working day in the future time period includes at least one sub-future time period, each sub-future time period corresponds to a target temperature-sensitive load, and d is an integer greater than c.

9. An electronic device, characterized in that: include: A processor and a memory, the processor is connected to the memory, the memory is used to store a computer program, and the processor is used to execute the computer program stored in the memory, so that the electronic device performs the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method according to any one of claims 1 to 7.

Images