Demand response potential prediction method and device, computer device and storage medium

Abstract

The application relates to a demand response potential prediction method and device, computer equipment, a storage medium and a computer program product. A demand response potential prediction request is detected, unit degree electricity production value, load peak-valley distribution coefficient and load interruptible coefficient of a to-be-predicted object are obtained from a power database, a preset rule table is inquired according to respective grades corresponding to the above coefficients, an elasticity coefficient matrix is determined according to obtained self-elasticity coefficients and mutual-elasticity coefficients, and new electricity demand of the to-be-predicted object in a preset time period in response to the updated electricity cost is determined according to the elasticity coefficient matrix, original electricity demand, original electricity cost, updated electricity cost and electricity time period quantity of the to-be-predicted object in the preset time period. Compared with traditional prediction by physical modeling, the scheme predicts the demand response potential based on unit degree electricity production value, peak-valley load coefficient, load interruptible coefficient and elasticity coefficient matrix, and the prediction complexity is reduced.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and in particular to a demand response potential prediction method, apparatus, computer equipment, storage medium, and computer program product. Background Art

[0002] As the proportion of renewable energy generation in the power system continues to increase, the power grid faces insufficient periodic peak-shaving capacity. Further tapping peak-shaving potential is needed to enhance the system's regulation capabilities and promote balanced electricity supply and demand. Demand response is an effective load management tool in the power industry. By appropriately raising electricity prices or providing incentives during peak demand periods and reducing prices during off-peak demand periods, consumers are encouraged to shift their electricity consumption, achieving the goal of shifting peak demand and filling valleys. Demand response can reduce and shift load during peak periods, effectively absorbing distributed generation and reducing grid operating costs. Assessing demand response potential is fundamental to developing scientific demand-side management measures and evaluating the effectiveness of demand response. It helps power companies and load managers determine the expected scale and source of demand response resources, thereby assisting in developing demand response development goals and strategies. Currently, demand response potential is typically estimated by performing physical modeling of various power devices. However, predicting demand response potential through physical modeling increases the complexity of the modeling process.

[0003] Therefore, the current demand response potential prediction methods have the disadvantage of high complexity. Summary of the Invention

[0004] Based on this, it is necessary to provide a demand response potential prediction method, device, computer equipment, computer-readable storage medium and computer program product that can reduce the complexity of prediction to address the above technical problems.

[0005] In a first aspect, the present application provides a method for predicting demand response potential, the method comprising:

[0006] In response to a demand response potential prediction request, query and obtain the unit kilowatt-hour electricity output value, load peak-valley distribution coefficient, and load interruptibility coefficient corresponding to the object to be predicted from the power database; the unit kilowatt-hour electricity output value represents the impact of electricity cost on the electricity consumption of the object to be predicted; the load peak-valley distribution coefficient represents the distribution characteristics of the electricity consumption of the object to be predicted during the electricity consumption period; and the load interruptibility coefficient represents the degree of impact of interrupting electricity consumption on the production of the object to be predicted;

[0007] According to the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient, a preset rule table is queried to obtain the corresponding self-elasticity coefficient and mutual elasticity coefficient; the preset rule table includes the corresponding relationship between the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient and the self-elasticity coefficient and the mutual elasticity coefficient; the self-elasticity coefficient represents the corresponding relationship between the electricity consumption of the object to be predicted in each time period and the change in electricity cost in the current time period; the mutual elasticity coefficient represents the corresponding relationship between the electricity consumption of the object to be predicted in each time period and the change in electricity cost in other time periods;

[0008] determining an elasticity coefficient matrix according to the self-elasticity coefficient and the mutual elasticity coefficient, and determining a new electricity demand of the object to be predicted in the preset time period in response to the updated electricity cost according to the elasticity coefficient matrix, the original electricity demand of the object to be predicted in the preset time period, the original electricity cost of the object to be predicted in the preset time period, the updated electricity cost corresponding to the object to be predicted in the preset time period, and the number of electricity time periods;

[0009] The demand response potential of the object to be predicted is determined according to the difference between the new electricity demand and the original electricity demand.

[0010] In one embodiment, obtaining the unit kilowatt-hour electricity output value, the load peak-valley distribution coefficient, and the load interruptibility coefficient corresponding to the object to be predicted includes:

[0011] Obtaining the annual output value increase and annual electricity consumption of the object to be predicted within a historical time period, and obtaining the total electricity consumption of the object to be predicted within a preset time period, wherein the preset time period includes a peak electricity consumption period, a valley electricity consumption period, and a level electricity consumption period;

[0012] Determine the unit kilowatt-hour output value of the object to be predicted based on the annual output value increase and annual electricity consumption in the historical time period;

[0013] Obtaining a first power consumption during a peak period, a second power consumption during a valley period, and the number of power consumption data points during the peak period, the valley period, and the level period from the total power consumption during the preset time period, and obtaining a load peak-valley distribution coefficient based on the first power consumption, the second power consumption, and the number of power consumption data points;

[0014] A load interruptibility coefficient corresponding to the object to be predicted is determined according to the power consumption behavior of the object to be predicted.

[0015] In one embodiment, before obtaining the first power consumption during the peak period, the second power consumption during the valley period, and the number of power consumption data points during the peak period, the valley period, and the level period, of the total power consumption during the preset time period, the method further includes:

[0016] Obtaining the region where the object to be predicted is located, and determining a division strategy for the preset time period into peak power consumption period, valley power consumption period, and level power consumption period according to the power consumption cost strategy corresponding to the region;

[0017] The peak power consumption period, the valley power consumption period and the level power consumption period within the preset time period are determined according to the division strategy.

[0018] In one embodiment, determining the division strategy of the preset time period into peak power consumption period, valley power consumption period and level power consumption period according to the power consumption cost strategy corresponding to the region includes:

[0019] If the electricity cost strategy is a time-of-use electricity price strategy, a strategy for dividing the preset time period into a peak electricity consumption period, a valley electricity consumption period, and a level electricity consumption period is determined according to the time-of-use electricity price strategy;

[0020] If the electricity cost strategy is not a time-of-use electricity price strategy, a division strategy of the preset time period into peak electricity consumption period, valley electricity consumption period and level electricity consumption period is determined according to a typical daily electricity consumption curve.

[0021] In one embodiment, querying a preset rule table according to the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient to obtain the corresponding self-elasticity coefficient and mutual elasticity coefficient includes:

[0022] Inputting the unit kilowatt-hour electricity output value into a first membership function to obtain a plurality of first probabilities of a plurality of unit kilowatt-hour electricity output value levels corresponding to the unit kilowatt-hour electricity output value output by the first membership function;

[0023] Inputting the load peak-valley distribution coefficient into a second membership function to obtain a plurality of second probabilities of a plurality of load peak-valley distribution coefficient levels corresponding to the load peak-valley distribution coefficient output by the second membership function;

[0024] Inputting the load interruptibility coefficient into a third membership function to obtain a plurality of third probabilities of a plurality of load interruptibility coefficient levels corresponding to the load interruptibility coefficient output by the third membership function;

[0025] According to the multiple unit kilowatt-hour electricity production value levels and the corresponding multiple first probabilities, the multiple load peak-valley distribution coefficient levels and the corresponding multiple second probabilities, and the multiple load interruptibility coefficient levels and the corresponding multiple third probabilities, a preset rule table is searched to obtain multiple fourth probabilities of multiple self-elasticity coefficient levels and multiple fifth probabilities of multiple mutual elasticity coefficient levels under the conditions of each group of unit kilowatt-hour electricity production value levels, load peak-valley distribution coefficient levels, and load interruptibility coefficient levels;

[0026] performing a logical AND operation on the plurality of fourth probabilities and the plurality of fifth probabilities, respectively, to determine at least one target self-elasticity coefficient level corresponding to the fourth probability of at least one target having the greatest probability corresponding to the object to be predicted, and to determine at least one target mutual elasticity coefficient level corresponding to the fifth probability of at least one target having the greatest probability corresponding to the object to be predicted;

[0027] Defuzzification is performed on the at least one target self-elasticity coefficient level and the at least one target mutual-elasticity coefficient level respectively according to the center of gravity method to obtain the self-elasticity coefficient and the mutual-elasticity coefficient of the object to be predicted.

[0028] In one embodiment, determining an elasticity coefficient matrix based on the self-elasticity coefficient and the mutual elasticity coefficient, and determining a new electricity demand of the object to be predicted in the preset time period in response to the updated electricity cost based on the elasticity coefficient matrix, the original electricity demand of the object to be predicted in the preset time period, the original electricity cost of the object to be predicted in the preset time period, the updated electricity cost corresponding to the object to be predicted in the preset time period, and the number of electricity time periods, includes:

[0029] Determine, based on the self-elasticity coefficient, a first coefficient between valley periods, a second coefficient between level periods, and a third coefficient between peak periods within the preset time period;

[0030] Determining, based on the mutual elasticity coefficient, a fourth coefficient between the peak power consumption period and the valley power consumption period, a fifth coefficient between the peak power consumption period and the level power consumption period, and a sixth coefficient between the level power consumption period and the valley power consumption period within the preset time period;

[0031] Obtaining the elastic coefficient matrix according to the first coefficient, the second coefficient, the third coefficient, the fourth coefficient, the fifth coefficient and the sixth coefficient;

[0032] Obtain the product of the elasticity coefficient matrix, the original electricity demand, the electricity cost change obtained by the difference between the original electricity cost and the updated electricity cost, and the ratio of the original electricity cost, and determine the new electricity load change of the object to be predicted in response to the updated electricity cost within the preset time period based on the ratio of the product to the number of electricity time periods, and then add the load change to the original electricity demand to obtain the updated electricity demand.

[0033] In a second aspect, the present application provides a demand response potential prediction device, the device comprising:

[0034] A response module is configured to query and obtain, from a power database, a unit kilowatt-hour electricity output value, a load peak-valley distribution coefficient, and a load interruptibility coefficient corresponding to an object to be predicted in response to a demand response potential prediction request; the unit kilowatt-hour electricity output value represents the impact of electricity cost on the electricity consumption of the object to be predicted; the load peak-valley distribution coefficient represents the distribution characteristics of the electricity consumption of the object to be predicted during the electricity consumption period; and the load interruptibility coefficient represents the degree of impact of interrupting electricity consumption on the production of the object to be predicted;

[0035] A query module is configured to query a preset rule table based on the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient to obtain corresponding self-elasticity coefficients and mutual elasticity coefficients; the preset rule table includes a correspondence between the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient and the self-elasticity coefficient and the mutual elasticity coefficient; the self-elasticity coefficient represents the correspondence between the electricity consumption of the object to be predicted in each time period and the change in electricity cost in the current time period; the mutual elasticity coefficient represents the correspondence between the electricity consumption of the object to be predicted in each time period and the change in electricity cost in other time periods;

[0036] a determination module, configured to determine an elasticity coefficient matrix based on the self-elasticity coefficient and the mutual elasticity coefficient, and determine a new electricity demand of the object to be predicted within the preset time period in response to the updated electricity cost based on the elasticity coefficient matrix, the original electricity demand of the object to be predicted within the preset time period, the original electricity cost of the object to be predicted within the preset time period, the updated electricity cost corresponding to the object to be predicted within the preset time period, and the number of electricity time periods;

[0037] The prediction module is used to determine the demand response potential of the object to be predicted based on the difference between the new electricity demand and the original electricity demand.

[0038] In a third aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0039] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the above method when executed by a processor.

[0040] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which implements the steps of the above method when executed by a processor.

[0041] The above-mentioned demand response potential prediction method, apparatus, computer device, storage medium, and computer program product obtain the unit kilowatt-hour power output value, load peak-valley distribution coefficient, and load interruptibility coefficient of the object to be predicted from the power database when a demand response potential prediction request is detected. A preset rule table is then queried based on the level corresponding to the unit kilowatt-hour power output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient to obtain the corresponding self-elasticity coefficient and mutual elasticity coefficient. The elasticity coefficient matrix is ​​determined based on the self-elasticity coefficient and the mutual elasticity coefficient. Furthermore, the new electricity demand of the object to be predicted in response to the updated electricity cost within the preset time period is determined based on the elasticity coefficient matrix, the original electricity demand of the object to be predicted within the preset time period, the original electricity cost of the object to be predicted within the preset time period, the updated electricity cost corresponding to the object to be predicted within the preset time period, and the number of electricity time periods. Compared to traditional prediction methods based on physical modeling, this solution reduces prediction complexity by predicting demand response potential based on unit kilowatt-hour power output value, peak-valley load factor, load interruptibility coefficient, and elasticity coefficient matrix. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a diagram of an application environment of a demand response potential prediction method in one embodiment;

[0043] Figure 2 Schematic diagram of a flow chart of a demand response potential prediction method in one embodiment;

[0044] Figure 3 A schematic flow chart of a demand response potential prediction method in another embodiment;

[0045] Figure 4 This is a schematic diagram of an interface for a demand response potential prediction step in one embodiment;

[0046] Figure 5 This is a schematic diagram of an interface for a demand response potential prediction step in one embodiment;

[0047] Figure 6is a structural block diagram of a demand response potential prediction device in one embodiment;

[0048] Figure 7 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0050] The demand response potential prediction method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. When receiving a demand response potential prediction request, the terminal 102 can obtain the unit kilowatt-hour electricity output value, load peak and valley distribution coefficient and load interruptibility coefficient corresponding to the prediction object from the database of the server 104, so that the terminal 102 can determine the corresponding elasticity coefficient matrix based on these coefficients, and then obtain the demand response potential with the prediction object. Among them, the terminal 102 can be but is not limited to various personal computers, laptops, smart phones and tablets. The server 104 can be implemented as an independent server or a server cluster consisting of multiple servers.

[0051] In one embodiment, Figure 2 As shown in the figure, a demand response potential prediction method is provided, which is applied to Figure 1 The following steps are used as an example to illustrate the terminal in the figure:

[0052] Step S202, in response to the demand response potential prediction request, query and obtain the unit kilowatt-hour electricity output value, load peak-valley distribution coefficient and load interruptibility coefficient corresponding to the object to be predicted from the power database; the unit kilowatt-hour electricity output value represents the impact of electricity cost on the electricity consumption of the object to be predicted; the load peak-valley distribution coefficient represents the impact of electricity time distribution on the electricity consumption of the object to be predicted; the load interruptibility coefficient represents the degree of impact of interrupting electricity consumption on the production of the object to be predicted.

[0053] Demand response, short for electricity demand response, refers to the short-term behavior of electricity users who, upon receiving a direct compensation notice from the power supplier inducing load reduction or a signal of rising electricity costs, change their established electricity usage patterns to reduce or shift their electricity consumption during a specific period. Demand response potential refers to a user's ability to participate in a demand response plan to adjust their load, including both increasing and reducing load. The target for which demand response potential prediction is required can be an industrial user, such as a metal manufacturer or textile mill. The terminal can predict the demand response potential of the target. This prediction can be triggered by a command. Upon detecting a user-triggered demand response potential prediction request, the terminal can query the power database and obtain the unit kilowatt-hour electricity output value, load peak-valley distribution coefficient, and load interruptibility coefficient corresponding to the target. The unit kilowatt-hour electricity output value represents the impact of electricity costs on the target's electricity consumption; the load peak-valley distribution coefficient represents the impact of the electricity consumption time distribution on the target's electricity consumption; and the load interruptibility coefficient represents the degree to which interrupting electricity consumption will affect the target. The above-mentioned unit kilowatt-hour electricity output value can be calculated based on the annual output value increase and annual electricity consumption of the object to be predicted; the above-mentioned load peak-valley distribution coefficient can be obtained based on the electricity consumption of the object to be predicted during the peak and valley periods; and the above-mentioned load interruptibility coefficient can be obtained based on the electricity consumption behavior of the object to be predicted.

[0054] Specifically, a start button for demand response potential prediction may be provided in the display device of the above-mentioned terminal, and the user may click the start button in the display device of the terminal, so that the terminal may receive a demand response potential prediction request, and query the power database to obtain information corresponding to the object to be predicted. Price-based demand response means that as electricity prices change dynamically, electricity users adjust their electricity usage behavior and demand in order to achieve the effect of reducing electricity expenditure. Price-based demand response mainly includes time-of-use electricity prices and peak electricity prices. The time-of-use electricity price scheme refers to dividing the 24 hours of a day into several time periods according to the system operation status, and charging electricity fees for each time period according to different charging standards to change the user's electricity usage behavior.

[0055] Step S204, query the preset rule table according to the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient to obtain the corresponding self-elasticity coefficient and mutual elasticity coefficient; the preset rule table includes the corresponding relationship between the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient and the self-elasticity coefficient and the mutual elasticity coefficient; the self-elasticity coefficient represents the corresponding relationship between the electricity consumption of the object to be predicted in each time period and the change in electricity cost in the current time period; the mutual elasticity coefficient represents the corresponding relationship between the electricity consumption of the object to be predicted in each time period and the change in electricity cost in other time periods.

[0056] After obtaining the aforementioned unit kilowatt-hour electricity output value, load peak-valley distribution coefficient, and load interruptibility coefficient, the terminal can obtain the levels of the aforementioned unit kilowatt-hour electricity output value, load peak-valley distribution coefficient, and load interruptibility coefficient, respectively, and obtain the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient. These levels can have multiple levels. Taking the unit kilowatt-hour electricity output value as an example, the corresponding level can be low, medium, and high. Each level can be determined by inputting the aforementioned unit kilowatt-hour electricity output value, load peak-valley distribution coefficient, and load interruptibility coefficient into their respective corresponding membership functions, and then determining the probability value of belonging to each level based on the membership function output. After obtaining the level corresponding to the aforementioned unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient, the terminal can query a preset rule table based on each level to obtain the corresponding self-elasticity coefficient and mutual-elasticity coefficient. The preset rule table includes the corresponding relationship between the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient and the self-elasticity coefficient and mutual-elasticity coefficient. The self-elasticity coefficient represents the corresponding relationship between the power consumption of the object to be predicted in each time period and the change in the power cost in the current time period; the mutual elasticity coefficient represents the corresponding relationship between the power consumption of the object to be predicted in each time period and the change in the power cost in other time periods.

[0057] Specifically, the terminal can combine the multiple levels corresponding to the aforementioned unit kilowatt-hour electricity output value, the multiple levels corresponding to the load peak-valley distribution coefficient, and the multiple levels corresponding to the load interruptibility coefficient to obtain multiple level conditions, each of which includes a level corresponding to the unit kilowatt-hour electricity output value, a level corresponding to the load peak-valley distribution coefficient, and a level corresponding to the load interruptibility coefficient. The terminal can then determine the corresponding self-elasticity coefficient level and mutual-elasticity coefficient level under each level condition to form a preset rule table. Thus, the terminal can form a level condition based on the level corresponding to the unit kilowatt-hour electricity output value of the object to be predicted, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient, query the preset rule table, and obtain the corresponding self-elasticity coefficient level and mutual-elasticity coefficient level. The terminal can then convert the self-elasticity coefficient level and mutual-elasticity coefficient level using the center of gravity method to obtain the self-elasticity coefficient and mutual-elasticity coefficient corresponding to the object to be predicted.

[0058] Step S206, determine the elasticity coefficient matrix based on the self-elasticity coefficient and the mutual elasticity coefficient, and determine the new electricity demand of the object to be predicted in the preset time period in response to the updated electricity cost based on the elasticity coefficient matrix, the original electricity demand of the object to be predicted in the preset time period, the original electricity cost of the object to be predicted in the preset time period, the updated electricity cost corresponding to the object to be predicted in the preset time period, and the number of electricity time periods.

[0059] After obtaining the self-elasticity coefficient and mutual elasticity coefficient, the terminal can determine an elasticity coefficient matrix based on the self-elasticity coefficient and mutual elasticity coefficient. The self-elasticity coefficient and mutual elasticity coefficient represent the impact of price changes during electricity consumption periods on electricity demand, including the impact of price changes during electricity consumption periods on electricity demand during the current period and the impact of price changes on electricity demand during other periods. Based on the self-elasticity coefficient and mutual elasticity coefficient, the terminal can determine the values ​​for comparisons between various time periods in the elasticity coefficient matrix, including comparisons between valley, level, and peak periods, thereby forming an elasticity coefficient matrix.

[0060] After obtaining the elasticity coefficient matrix, the terminal can determine the new electricity demand of the object to be predicted in response to the updated electricity cost within the preset time period based on the elasticity coefficient matrix, the original electricity demand of the object to be predicted in the preset time period, the original electricity cost of the object to be predicted in the preset time period, the updated electricity cost corresponding to the object to be predicted in the preset time period, and the number of electricity time periods. For example, the terminal determines the electricity demand curve of the object to be predicted in the preset time period based on the original electricity demand and electricity cost, and determines the new electricity demand curve of the object to be predicted in response to the updated electricity cost within the preset time period based on the updated electricity cost and the number of electricity time periods. The original electricity cost can be the electricity cost at a certain time when time-of-use electricity prices are not used for calculation, and the updated electricity cost can be the electricity cost at a certain time when time-of-use electricity prices are used for calculation. The number of electricity time periods can be the number of time periods within a day that require electricity price adjustment after the adoption of time-of-use electricity prices. The original electricity demand can be the load of the object to be predicted when it does not adjust its electricity consumption in response to the change in electricity cost. The new electricity demand can be the load when the object to be predicted responds to the change in electricity cost and adjusts the electricity consumption.

[0061] Step S208: determining the demand response potential of the object to be predicted based on the difference between the new electricity demand and the original electricity demand.

[0062] In which, the above-mentioned preset time period can be a 24-hour time period of a day, the new electricity demand can include the new electricity demand at each time of the day, and the original electricity demand can include the original electricity demand at each time of the day, then the terminal can form an original electricity curve and a new electricity curve according to the original electricity demand at each time and the new electricity demand at each time, respectively, and the terminal can determine the demand response potential of the object to be predicted based on the difference between the new electricity demand and the original electricity demand. For example, the terminal can determine the difference in electricity demand by comparing the original electricity curve and the new electricity curve, and determine the demand response potential of the object to be predicted based on the difference in electricity demand. For example, the greater the difference in electricity demand, the greater the demand response potential of the object to be predicted. The terminal can also perform potential prediction of the demand resources of the object to be predicted by using a potential prediction model of price-based demand response. The price-based demand response potential prediction model is constructed based on the elasticity coefficient matrix, the original electricity demand of the forecasted object within a preset time period, the original electricity cost of the forecasted object within the preset time period, the updated electricity cost corresponding to the forecasted object within the preset time period, and the number of electricity time periods. Thus, the terminal can obtain the demand response potential of the forecasted object based on the change in electricity cost.

[0063] In the above-mentioned demand response potential prediction method, upon detecting a demand response potential prediction request, the unit kilowatt-hour electricity output value, the load peak-valley distribution coefficient, and the load interruptibility coefficient of the object to be predicted are obtained from the power database. A preset rule table is then queried based on the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient to obtain the corresponding self-elasticity coefficient and mutual elasticity coefficient. The elasticity coefficient matrix is ​​then determined based on the self-elasticity coefficient and the mutual elasticity coefficient. The new electricity demand of the object to be predicted in response to the updated electricity cost within the preset time period is determined based on the elasticity coefficient matrix, the original electricity demand of the object to be predicted within the preset time period, the original electricity cost of the object to be predicted within the preset time period, the updated electricity cost corresponding to the object to be predicted within the preset time period, and the number of electricity time periods. Compared to traditional prediction methods based on physical modeling, this solution reduces prediction complexity by predicting demand response potential based on unit kilowatt-hour electricity output value, peak-valley load factor, load interruptibility coefficient, and elasticity coefficient matrix.

[0064] In one embodiment, the unit degree electricity production value, the load peak-valley distribution coefficient and the load interruptible coefficient corresponding to the object to be predicted are obtained, including: obtaining the annual production value increment and the annual electricity consumption of the object to be predicted in a historical period; and obtaining the total electricity consumption of the object to be predicted in a preset period, the preset period including a peak electricity consumption period, a valley electricity consumption period and a flat electricity consumption period; determining the unit degree electricity production value of the object to be predicted according to the annual production value increment and the annual electricity consumption in the historical period; obtaining the first electricity consumption in the peak electricity consumption period, the second electricity consumption in the valley electricity consumption period and the number of electricity consumption data points in the peak electricity consumption period, the valley electricity consumption period and the flat electricity consumption period in the total electricity consumption in the preset period, and obtaining the load peak-valley distribution coefficient according to the first electricity consumption, the second electricity consumption and the number of electricity consumption data points; determining the load interruptible coefficient corresponding to the object to be predicted according to the electricity consumption behavior of the object to be predicted.

[0065] In the embodiment, the terminal can obtain the unit degree electricity production value, the load peak-valley distribution coefficient and the load interruptible coefficient of the object to be predicted by different calculation methods. For the unit degree electricity production value, the terminal can obtain the annual production value increment and the annual electricity consumption of the object to be predicted in a historical period. The preset period includes a peak electricity consumption period, a valley electricity consumption period and a flat electricity consumption period, and the division strategy of the peak electricity consumption period, the valley electricity consumption period and the flat electricity consumption period can be different based on different electricity cost strategies. For example, in one embodiment, before obtaining the first electricity consumption in the peak electricity consumption period, the second electricity consumption in the valley electricity consumption period and the number of electricity consumption data points in the peak electricity consumption period, the valley electricity consumption period and the flat electricity consumption period in the total electricity consumption in the preset period, the terminal further includes: obtaining the region where the object to be predicted is located, determining the division strategy of the peak electricity consumption period, the valley electricity consumption period and the flat electricity consumption period in the preset period according to the electricity cost strategy corresponding to the region, and determining the peak electricity consumption period, the valley electricity consumption period and the flat electricity consumption period in the preset period according to the division strategy. In the embodiment, the terminal can detect the region where the object to be predicted is located and obtain the electricity cost strategy corresponding to the region, where the electricity cost can be the electricity price. The terminal can determine the division strategy of the peak electricity consumption period, the valley electricity consumption period and the flat electricity consumption period in the preset period according to the electricity cost strategy of the region, and determine the division strategy of the peak electricity consumption period, the valley electricity consumption period and the flat electricity consumption period in the preset period according to the division strategy.

[0066] The determination of the electricity cost strategy may specifically be a determination by the terminal as to whether a region has a time-of-use electricity price strategy. For example, in one embodiment, a strategy for dividing a preset time period into peak, valley, and level periods is determined based on the electricity cost strategy corresponding to the region, including: if the electricity cost strategy is a time-of-use electricity price strategy, the strategy for dividing the preset time period into peak, valley, and level periods is determined based on the time-of-use electricity price strategy; if the electricity cost strategy is not a time-of-use electricity price strategy, the strategy for dividing the preset time period into peak, valley, and level periods is determined based on a typical daily electricity consumption curve. In this embodiment, the terminal can determine whether there is a time-of-use electricity price strategy in the area where the above-mentioned object to be predicted is located. If the electricity cost strategy in the area where the object to be predicted is located is a time-of-use electricity price strategy, the terminal can determine the division strategy of the peak electricity consumption period, valley electricity consumption period and level electricity consumption period in the preset time period according to the time-of-use electricity price strategy, that is, the division strategy is related to the time-sharing method of the time-of-use electricity price strategy; if the above-mentioned electricity cost strategy is not a time-of-use electricity price strategy, the terminal can determine the division strategy of the peak electricity consumption period, valley electricity consumption period and level electricity consumption period in the above-mentioned preset time period according to the typical daily electricity consumption curve of the power industry, that is, the division strategy is related to the electricity consumption changes on a typical day.

[0067] The terminal can obtain the annual output value increase and annual electricity consumption of the above-mentioned target original object to be predicted in the historical time period; and obtain the total electricity consumption within the preset time period of the object to be predicted, the preset time period includes peak consumption period, valley consumption period and level consumption period and the distribution of each time period.

[0068] When the terminal obtains the output value increase and total electricity consumption of the object to be predicted within a preset time period, it is necessary to first determine the object to be predicted. For example, in one embodiment, obtaining the output value increase and total electricity consumption of the object to be predicted within a preset time period includes: obtaining multiple original objects to be predicted, and obtaining the original output value increase and original total electricity consumption of each original object to be predicted within the preset time period; sorting the multiple original objects to be predicted in descending order based on the original total electricity consumption, obtaining a target original object to be predicted whose ranking is less than a preset ranking threshold as the object to be predicted, and obtaining the original output value increase and original total electricity consumption corresponding to the target original object to be predicted as the output value increase and total electricity consumption of the object to be predicted. In this embodiment, the terminal can determine the object to be predicted through conditional screening. The terminal can obtain multiple original objects to be predicted and obtain the original output value increase and original total electricity consumption of each original object to be predicted within the preset time period. The terminal can sort the multiple original objects to be predicted in descending order based on the original total electricity consumption, and obtain the target original object to be predicted whose ranking is less than the preset ranking threshold as the object to be predicted that participates in the demand response potential prediction. The terminal may obtain the original output value increase and the original total electricity consumption corresponding to the above-mentioned target original object to be predicted as the output value increase and the total electricity consumption of the object to be predicted.

[0069] Specifically, the terminal can classify electricity users by industry and collect peak-load data for each industry within a preset region. This data is represented by 96 points, representing one point every 15 minutes throughout the day. The interval for acquiring this data can be determined based on actual needs. The terminal can also obtain annual load data and related economic statistics for each industry, and can also determine the measurement range of the price elasticity coefficient matrix for different regions.

[0070] After obtaining the annual output value increase and annual electricity consumption of the historical time period of the object to be predicted, the terminal can determine the unit kilowatt-hour output value of the object to be predicted based on the annual output value increase and total electricity consumption. Specifically, the above unit electricity output value calculation formula can be as follows:

[0071] Among them, α is the unit electricity output value, P gdp Q is the annual output value increase of the object to be predicted in the historical period, l The annual electricity consumption of the forecasted object over the historical time period. The unit kilowatt-hour output value α is used to determine the industry's sensitivity to electricity costs. A low unit kilowatt-hour output value α indicates a high sensitivity to electricity prices. This means that price fluctuations significantly impact the industry's costs and revenues, and thus its electricity consumption patterns. This indicates a high demand response potential. A high unit kilowatt-hour output value α indicates a low sensitivity to electricity prices and a low demand response potential.

[0072] For the load peak-valley distribution coefficient, the terminal may obtain a first power consumption during a peak period, a second power consumption during a valley period, and the number of power consumption data points during the peak and valley periods within a preset time period, and determine the load peak-valley distribution coefficient based on the first power consumption, the second power consumption, and the number of power consumption data points. The power consumption data points may be time points set at preset time intervals within the preset time period. Specifically, the preset time period may be a day, and the terminal may set a power consumption data point every fifteen minutes. Thus, a day will contain 96 power consumption data points, and the peak period, valley period, and level period will have corresponding power consumption data points. The object to be predicted may be within a preset area, and the terminal may adopt a time-of-use electricity pricing scheme for the preset area. Based on the time-of-use electricity pricing scheme or the load distribution on the maximum load day within the preset area, the terminal may divide the preset time period, such as a day, into three time periods: a peak period, a valley period, and a level period. High electricity price periods correspond to peak periods, level electricity price periods correspond to level periods, and low electricity price periods correspond to valley periods. The calculation formula of the above load peak-valley distribution coefficient can be shown as follows:

[0073] Where β is the load peak-valley distribution coefficient, is the total load during peak hours, is the total load during the valley period, m is the number of peak periods, also known as the number of electricity consumption data points during the peak period, and n is the number of valley periods, also known as the number of electricity consumption data points during the valley period.

[0074] The terminal can judge the industry's willingness and degree of participation in price-based demand response based on the load peak-valley distribution coefficient β. When the load peak-valley distribution coefficient β is greater than 1, it means that the unit load of the industry during peak hours is greater than the unit load during valley hours, and the industry has strong load reduction and load shifting potential. The higher the load peak-valley distribution coefficient β of the industry, the greater the potential for load reduction and load shifting of the industry. When the peak-valley load correlation coefficient β is less than 1, it means that the unit load of the industry during valley hours is greater than the unit load during peak hours, and the potential for load reduction or load shifting of the industry is considered low.

[0075] Regarding the load interruptibility coefficient, the terminal can determine the load interruptibility coefficient corresponding to the target based on its electricity usage behavior. This electricity usage behavior can be related to how the target uses electricity during production, and it represents the target's dependence on electricity. Specifically, the terminal can determine the load interruptibility coefficient θ by analyzing the electricity usage characteristics and logic of the industry to which the target belongs. This refers to collecting relevant data, conducting research, and analyzing the load usage patterns, electricity usage patterns, and industry characteristics of the industry. For example, in the chemical materials production industry, the primary electricity load comes from processing furnaces and production lines. Interrupting load or operating at low load presents significant safety risks, making it difficult to resume production and creating safety hazards that could potentially cause accidents to the entire production system and personnel. Due to this electricity usage characteristic of the chemical materials production industry, it can be considered that the industry has low load shifting and load interruption capabilities. The load interruptibility coefficient θ is defined as 0 when the industry has no load shifting capability at all, and as 1 when the industry has strong load shifting capability. The load transfer capability indicates whether the object to be predicted can produce through other energy sources when power is cut off.

[0076] Through the above embodiments, the terminal can determine the unit kilowatt-hour electricity output value, load peak-valley distribution coefficient and load interruptibility coefficient corresponding to the object to be predicted based on multiple methods, so that the terminal can predict the demand response potential of the object to be predicted based on the unit kilowatt-hour electricity output value, load peak-valley distribution coefficient and load interruptibility coefficient, thereby reducing the complexity of the demand response potential prediction.

[0077] In one embodiment, a preset rule table is queried according to the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient to obtain the corresponding self-elasticity coefficient and mutual elasticity coefficient, including: inputting the unit kilowatt-hour electricity output value into a first membership function to obtain a plurality of first probabilities of a plurality of unit kilowatt-hour electricity output value levels corresponding to the unit kilowatt-hour electricity output value output by the first membership function; inputting the load peak-valley distribution coefficient into a second membership function to obtain a plurality of second probabilities of a plurality of load peak-valley distribution coefficient levels corresponding to the load peak-valley distribution coefficient output by the second membership function; inputting the load interruptibility coefficient into a third membership function to obtain a plurality of third probabilities of a plurality of load interruptibility coefficient levels corresponding to the load interruptibility coefficient output by the third membership function; and and the corresponding multiple second probabilities and multiple load interruptible coefficient levels and the corresponding multiple third probabilities query the preset rule table to obtain multiple fourth probabilities of multiple self-elasticity coefficient levels and multiple fifth probabilities of multiple mutual elasticity coefficient levels under the conditions of each group of unit kilowatt-hour power output value level, load peak-valley distribution coefficient level and load interruptible coefficient level; perform logical AND operation on the multiple fourth probabilities and the multiple fifth probabilities respectively to determine at least one target self-elasticity coefficient level corresponding to at least one target fourth probability with the largest probability corresponding to the object to be predicted, and determine at least one target mutual elasticity coefficient level corresponding to at least one target fifth probability with the largest probability corresponding to the object to be predicted; defuzzify the at least one target self-elasticity coefficient level and the at least one target mutual elasticity coefficient level according to the center of gravity method to obtain the self-elasticity coefficient and mutual elasticity coefficient of the object to be predicted.

[0078] In this embodiment, the above-mentioned unit kilowatt-hour electricity output value, load peak-valley distribution coefficient, and load interruptibility coefficient can each have corresponding levels. The terminal can determine the level of each parameter using the corresponding membership function. The terminal can input the unit kilowatt-hour electricity output value into a first membership function to obtain multiple first probabilities for multiple unit kilowatt-hour electricity output value levels corresponding to the unit kilowatt-hour electricity output value output by the first membership function. That is, there can be multiple unit kilowatt-hour electricity output value levels. Based on the first membership function, the terminal can obtain a first probability that the unit kilowatt-hour electricity output value belongs to each level, thereby obtaining multiple first probabilities. The terminal can also input the load peak-valley distribution coefficient into a second membership function to obtain multiple second probabilities for multiple load peak-valley distribution coefficient levels corresponding to the load peak-valley distribution coefficient output by the second membership function. That is, there can be multiple load peak-valley distribution coefficient levels. Based on the second membership function, the terminal can obtain a second probability that the load peak-valley distribution coefficient belongs to each level, thereby obtaining multiple second probabilities. The terminal can also input the load interruptibility coefficient into a third membership function to obtain multiple third probabilities for multiple load interruptibility coefficient levels corresponding to the load interruptibility coefficient output by the third membership function. That is, there can be multiple load interruptibility coefficient levels, and the terminal can obtain the third probability that the load interruptibility coefficient belongs to each level based on the third membership function, thereby obtaining multiple third probabilities. Determination of each of the above levels can be a fuzzy process. The terminal can form a set of judgment conditions based on a unit kilowatt-hour power output value level, a load peak-valley distribution coefficient level, and a load interruptibility coefficient level. The terminal can query a preset rule table based on the multiple unit kilowatt-hour power output value levels and the corresponding multiple first probabilities, the multiple load peak-valley distribution coefficient levels and the corresponding multiple second probabilities, and the multiple load interruptibility coefficient levels and the corresponding multiple third probabilities to obtain multiple fourth probabilities for multiple self-elasticity coefficient levels and multiple fifth probabilities for multiple mutual-elasticity coefficient levels under each set of unit kilowatt-hour power output value level, load peak-valley distribution coefficient level, and load interruptibility coefficient level conditions. Thus, the terminal can perform logical AND operations on the multiple fourth probabilities and the multiple fifth probabilities, respectively, to determine at least one target self-elasticity coefficient level corresponding to the at least one target fourth probability with the highest probability corresponding to the object to be predicted, and to determine at least one target mutual elasticity coefficient level corresponding to the at least one target fifth probability with the highest probability corresponding to the object to be predicted. Since the multiple probability values ​​may be identical, the target fourth probability and the target fifth probability can each be at least one. After determining the at least one target self-elasticity coefficient level and the at least one target mutual elasticity coefficient level with the highest probability, the terminal can defuzzify the at least one target self-elasticity coefficient level and the at least one target mutual elasticity coefficient level using the center of gravity method to obtain the self-elasticity coefficient and mutual elasticity coefficient of the object to be predicted.

[0079] Specifically, the process of obtaining the grade can be a fuzzification process. For a precise input value, the corresponding linguistic variable value, i.e., the fuzzy variable value, is found. This linguistic variable value is then described in natural language. This process is referred to as fuzzification. The corresponding degree of membership is obtained based on the appropriate natural language value, and this natural language variable becomes a fuzzy subset. The terminal can be equipped with a fuzzy controller. The terminal can input the unit kilowatt-hour power output value α, the load peak-valley distribution coefficient β, and the load interruptibility coefficient θ into the fuzzy controller, each described using three fuzzy sets. The grade descriptions of each parameter can be as follows: unit kilowatt-hour power output value: low (SA), medium (MA), and high (LA); load peak-valley distribution coefficient: low (SB), medium (MB), and high (LB); load interruptibility coefficient: low (SC), medium (MC), and high (LC). For the self-elasticity coefficient e1 and the mutual elasticity coefficient e2, the terminal can determine the level of the self-elasticity coefficient e1 and the level of the mutual elasticity coefficient e2 by the following description: Self-elasticity coefficient e1: very low (VS1), low (S1), medium (M1), high (L1), very high (VL1); Mutual elasticity coefficient e2: very low (VS2), low (S2), medium (M2), high (L2), very high (VL2). The terminal can obtain the probability that each of the above parameters belongs to each level, and use a unit kilowatt-hour power output value, a load peak-valley distribution coefficient, and a load interruptibility coefficient as a set of judgment conditions to determine the elasticity coefficient and its probability under each set of conditions, forming a preset rule table. The table can be as follows:

[0080]

[0081] The table above describes 27 fuzzy rules. The first rule is: "If α is SA, β is SB AND θ is SC, THEN e1 is M1 AND e2 is VS2," which means, "If the unit kilowatt-hour electricity output value α is small, the load peak-valley distribution coefficient β is low, and the load interruptibility coefficient θ is small, then the self-elasticity coefficient e1 is medium, and the mutual elasticity coefficient e2 is very low." When determining the target self-elasticity coefficient and mutual elasticity coefficient, the terminal can first substitute the three coefficient indicators of the predicted object into the corresponding membership function and calculate the membership degree corresponding to each input indicator. Then, the terminal searches the fuzzy rule base for matching fuzzy rules based on the membership degree, determining the corresponding level of the coefficient and the probability of the level. The terminal can perform rule premise reasoning: Within each rule, the conclusion can be inferred from the premises through the "and" relationship between the rule premises. For example, if X is PA and Y is NA, then Z is VS, the membership degree of the premise is calculated using the minimum operation. The membership degree of the premise of this rule is min(μ PA(X) , μ NA(Y))μ PA(X) The membership degree representing X as PA can also be called probability. The terminal can also take the union of the membership degrees of all rule premises of the same type to obtain the total membership output of the fuzzy system. From this total membership output, the terminal can obtain at least one target self-elasticity coefficient level and at least one target mutual elasticity coefficient level with the maximum membership degree, i.e., the maximum probability value. After obtaining the at least one target self-elasticity coefficient level and at least one target mutual elasticity coefficient level, the terminal can defuzzify the at least one target self-elasticity coefficient level and at least one target mutual elasticity coefficient level using the area centroid method to obtain the self-elasticity coefficient and mutual elasticity coefficient of the object to be predicted. Furthermore, the terminal can obtain the price elasticity coefficient matrix E based on the self-elasticity coefficient and mutual elasticity coefficient.

[0082] Through the above embodiments, the terminal can determine the levels of various parameters corresponding to the object to be predicted based on the membership function, and determine the self-elasticity coefficient and the mutual elasticity coefficient based on the levels and the preset rule table, so that the terminal can construct an elasticity coefficient matrix based on the self-elasticity coefficient and the mutual elasticity coefficient, and predict the demand response potential of the object to be predicted, thereby reducing the prediction complexity.

[0083] In one embodiment, an elasticity coefficient matrix is ​​determined based on the self-elasticity coefficient and the mutual elasticity coefficient, and a new electricity demand of the object to be predicted in the preset time period in response to the updated electricity cost is determined based on the elasticity coefficient matrix, the original electricity demand of the object to be predicted in the preset time period, the original electricity cost of the object to be predicted in the preset time period, the updated electricity cost corresponding to the object to be predicted in the preset time period, and the number of electricity time periods, including: determining a first coefficient between a valley period and a valley period, a second coefficient between a level period and a level period, and a third coefficient between a peak period and a peak period within the preset time period based on the self-elasticity coefficient; determining a first coefficient between a valley period and a valley period, a second coefficient between a level period and a level period, and a third coefficient between a peak period and a peak period within the preset time period based on the mutual elasticity coefficient; determining a first coefficient between a valley period and a valley period, a second coefficient between a level period and a level period, and a third coefficient between a peak period and a peak period within the preset time period based on the mutual elasticity coefficient. The fourth coefficient between the peak period and the valley period, the fifth coefficient between the peak period and the level period, and the sixth coefficient between the level period and the valley period; according to the first coefficient, the second coefficient, the third coefficient, the fourth coefficient, the fifth coefficient and the sixth coefficient, an elasticity coefficient matrix is ​​obtained; the elasticity coefficient matrix, the original electricity demand, the product of the change in electricity cost obtained by the difference between the original electricity cost and the updated electricity cost and the ratio of the original electricity cost are obtained, and according to the ratio of the product to the number of electricity periods, the new electricity load change of the object to be predicted in response to the updated electricity cost within the preset time period is determined, and the updated electricity demand is obtained according to the sum of the new electricity load change and the original electricity demand.

[0084] In this embodiment, after obtaining the self-elasticity coefficient and mutual elasticity coefficient of the object to be predicted, the terminal can determine, based on the self-elasticity coefficient, a first coefficient between valley periods, a second coefficient between level periods, and a third coefficient between peak periods within a preset time period. Furthermore, based on the mutual elasticity coefficient, the terminal can determine, based on the mutual elasticity coefficient, a fourth coefficient between peak periods and valley periods, a fifth coefficient between peak periods and level periods, and a sixth coefficient between level periods and valley periods within the preset time period. That is, each elasticity coefficient represents the influence relationship between different power consumption periods. The terminal can obtain an elasticity coefficient matrix based on the first, second, third, fourth, fifth, and sixth coefficients. After the terminal obtains the elasticity coefficient matrix, it can obtain the product of the elasticity coefficient matrix, the original electricity demand, the electricity cost change obtained by the difference between the original electricity cost and the updated electricity cost, and the ratio of the original electricity cost, and determine the new electricity load change of the object to be predicted in response to the updated electricity cost within the preset time period based on the ratio of the product to the number of electricity time periods. Then, the load change is added to the original electricity demand to obtain the updated electricity demand.

[0085] Specifically, the elastic coefficient matrix E can be expressed as follows:

[0086] Among them, ε is the demand elasticity coefficient, which can also be called e. It is not only related to the price changes in this period, but also related to the price changes in other periods. The terminal can use the demand elasticity coefficient ε ii and cross elastic coefficient ε ij Indicates the part related to price changes in this period and the part related to price changes in other periods. ii Indicates the impact of electricity price changes in period i on demand in period i, and its value is usually negative; ε ij It represents the impact of electricity price changes in period j on demand in period i, and its value is usually positive. After the terminal obtains the above elasticity coefficient matrix, it can determine the new electricity demand based on the following formula:

[0087] Among them, Q is the original electricity demand corresponding to a certain moment, Q' is the electricity demand after demand response corresponding to a certain moment; P is the original electricity cost corresponding to a certain moment, ΔP is the change in the demand response electricity cost corresponding to a certain moment; E is the price elasticity coefficient matrix; n is the number of time periods divided each day.

[0088] Through the above embodiment, the terminal can determine the new electricity demand corresponding to the object to be predicted after the response electricity price is updated based on the elasticity coefficient matrix determined by the self-elasticity coefficient and the mutual elasticity coefficient. The terminal can predict the demand response of the object to be predicted based on the new electricity consumption and the original electricity consumption, thereby reducing the prediction complexity.

[0089] In this embodiment, after determining the new electricity demand of the target object in response to the change in electricity cost, the terminal can obtain the difference between the new electricity demand and the original electricity demand. Both the new electricity demand and the original electricity demand can include demands at multiple time points. The terminal can then construct an electricity demand curve based on the demands at multiple time points and determine the demand response potential of the target object based on the difference in the electricity demand curves.

[0090] Through this embodiment, the terminal can determine the demand response potential of the object to be predicted based on the difference between the new electricity demand and the original electricity demand, thereby reducing the complexity of prediction.

[0091] In one embodiment, the demand response potential of the object to be predicted is determined based on the difference between the new electricity demand and the original electricity demand, including: if the difference between the new electricity demand and the original electricity demand is greater than a preset electricity demand threshold, determining that the demand response potential of the object to be predicted at the current electricity cost is the first level; if the difference between the new electricity demand and the original electricity demand is less than or equal to the preset electricity demand threshold, determining that the demand response potential of the object to be predicted at the current electricity cost is the second level; wherein the first level is greater than the second level.

[0092] In this embodiment, after the terminal determines the new electricity demand of the object to be predicted in response to the change in electricity cost, it can obtain the difference between the new electricity demand and the original electricity demand. If the terminal detects that the difference between the new electricity demand and the original electricity demand is greater than the preset electricity demand threshold, the terminal can determine that the demand response potential of the object to be predicted under the current electricity cost is the first level. If the terminal detects that the difference between the new electricity demand and the original electricity demand is less than or equal to the preset electricity demand threshold, the terminal can determine that the demand response potential of the object to be predicted under the current electricity cost is the second level; wherein the first level is greater than the second level, the first level indicates that the demand response potential of the object to be predicted is large, and the second level indicates that the demand response potential of the object to be predicted is small. Wherein, the above-mentioned new electricity demand and the original electricity demand can both include demand at multiple time points, and the terminal can construct an electricity curve based on the demand at multiple time points, and determine the demand response potential of the object to be predicted based on the difference in the electricity curve.

[0093] Through this embodiment, the terminal can determine the demand response potential of the object to be predicted based on the difference between the new electricity demand and the original electricity demand, thereby reducing the complexity of prediction.

[0094] In one embodiment, Figure 3 As shown, Figure 3This is a flow chart of another embodiment of a demand response potential prediction method. In this embodiment, the terminal designs an index system, including three coefficient indicators: unit kilowatt-hour output value α, peak-valley load correlation coefficient β, and load interruptibility coefficient θ. The terminal first calculates the correlation coefficient indicators for the selected industry. By establishing a fuzzy control model, the mapping relationship between the index system and the elasticity coefficient matrix is ​​obtained. The demand elasticity matrix E is then quantitatively analyzed and obtained. Finally, the adjustable potential of demand-side resources based on time-of-use electricity prices is calculated using the above-mentioned new electricity demand formula. Specifically, the terminal first collects 96 points of electricity load data (one point every 15 minutes throughout the day) for each industry in a certain region on the day of maximum load, annual industry load data, and related economic statistics. The terminal also obtains the calculation range of the price elasticity coefficient matrix for different regions at home and abroad. The terminal can perform an initial screening of industries, selecting the selected industries based on the target of analysis. In this patent, the total daily load of each industry is ranked by its proportion of total electricity consumption in the society, and the industries with the highest electricity consumption in the society are selected as the targets for prediction. The terminal can use the above formula to calculate the unit kilowatt-hour output value α. This value is used to determine the industry's sensitivity to electricity costs. When the unit kilowatt-hour output value α is low, the industry is considered to be highly sensitive to electricity prices. This means that fluctuations in electricity prices have a significant impact on the industry's costs and revenues, and thus influence its electricity consumption patterns. This indicates that the industry has greater demand response potential. When the unit kilowatt-hour output value α is high, the industry is considered to be less sensitive to electricity prices and has less demand response potential. The terminal can also calculate the load peak-valley distribution coefficient β, which is used to assess the industry's willingness and degree of participation in price-based demand response. When the load peak-valley distribution coefficient β is greater than 1, it indicates that the industry's unit load during peak hours is greater than during off-peak hours, indicating that the industry has strong potential for load reduction and load shifting. The higher the load peak-valley distribution coefficient β, the greater the potential for load reduction and load shifting. When the peak-valley load correlation coefficient β is less than 1, it indicates that the industry's unit load during off-peak hours is greater than during peak hours, indicating that the industry has low potential for load reduction or load shifting. The terminal can also calculate the load interruptibility factor θ, which is determined by analyzing the industry's electricity usage characteristics and logic. Specifically, this factor is determined by collecting relevant data, conducting research, and analyzing the load power usage patterns, usage patterns, and industry characteristics of the industry. For example, in the chemical materials production industry, the primary power load comes from processing furnaces and production lines. Interrupting load or operating at low loads presents significant safety risks. Resuming production after a furnace shutdown is difficult and can easily create safety hazards, potentially causing accidents to the entire production system and personnel. Due to this electricity usage characteristic of the chemical materials production industry, it can be considered that the industry has a low capacity for load shifting and load interruption.It is defined that when the industry has no load transfer capability at all, the load interruptibility coefficient θ is 0; when the industry has a strong load transfer capability, the load interruptibility coefficient θ is 1.

[0095] The terminal can also describe the above coefficients in natural language, that is, determine the level of the above coefficients through fuzzy processing, and formulate a fuzzy rule table, that is, the above preset rule table, and determine the elasticity coefficient matrix E based on the preset rule table. Specifically, the terminal substitutes the three coefficient indicators calculated by each industry into the membership function, and then calculates the membership corresponding to each input indicator. Then, the matching fuzzy rules are searched for by the membership in the fuzzy rule base. And rule premise reasoning is performed, including: in each rule, the conclusions can be drawn between the premises of each rule through the "and" relationship. For example, the rule is IF Xis PA and Y is NA then Z is VS. Then, by taking the smaller operation, the membership of the premise is calculated. The membership of the premise of this rule is min(μ PA(X) , μ NA(Y) The terminal can take the union of the membership degrees of all the same type of rule premises to obtain the total membership output of the fuzzy system. The area centroid method is then used for defuzzification to obtain the price elasticity coefficient matrix E. Based on the demand elasticity coefficient matrix E obtained through quantitative analysis, the terminal can use the established price-based demand response potential calculation model to calculate the potential of demand-side resources. For example, the elasticity coefficient matrix E can be used to determine the new electricity demand of the object to be predicted. Thus, the terminal can determine the demand response potential of the object to be predicted based on the new electricity demand and the original electricity demand.

[0096] Through the above embodiment, the terminal reduces the prediction complexity by predicting the demand response potential based on the unit kilowatt-hour electricity output value, peak and valley load factor, load interruptibility factor and elasticity coefficient matrix.

[0097] This application also provides an application example. For example, a terminal can select the metal products and textile industries as case analysis samples, and based on the load data within a preset time period, calculate the adjustable potential of these industries. First, the important indicators of these key industries are calculated, and the indicator calculation results are shown in the following table.

[0098]

[0099] The terminal can determine the three fuzzy sets corresponding to each coefficient, i.e., three levels, according to the settings of the levels corresponding to the above-mentioned coefficients. The fuzzy subsets of the input and output of the fuzzy control system use triangular and trapezoidal membership functions. The fuzzy rules are set according to the above-mentioned fuzzy rule design table, i.e., the above-mentioned preset rule table. The terminal can match the fuzzy rules using the above-mentioned fuzzy rule design table and the fuzzy membership function. For example, given the specific values ​​of the three input variables, such as the unit kilowatt-hour power output value α is 10, the load peak-valley distribution coefficient β is 1, and the load interruptibility coefficient θ is 0.25, the corresponding memberships are:

[0100]

[0101] The terminal can substitute the above specific values ​​into the above preset rule table to obtain the following table:

[0102]

[0103] The terminal can obtain eight fuzzy rules from the above preset rule table and perform rule premise reasoning. Specifically, the terminal can perform logical AND operations between premises within the same rule to obtain the conclusions of each fuzzy rule and the membership degree of each fuzzy rule premise, as shown in the following table:

[0104]

[0105] The terminal can intersect the above two tables and obtain the fuzzy output of each rule through logical AND operation. Among them, the total membership result of the fuzzy system self-elasticity coefficient e1 is: μ agg (e1)=max{min(1 / 3, μ M1(e1) ),min(1 / 3,μ M1(e1) ),min(1 / 3,μ M1(e1) ),min(1 / 3,μ M1(e1) ),min(1 / 3,μ S1(e1) ),min(1 / 3,μ S1(e1) ),min(1 / 2,μ S1(e1) ),min(1 / 2,μ M1(e1) )}=max{min(1 / 2, μ S1(e1) ),min(1 / 2,μ M1(e1) )}. The total membership degree of the fuzzy system self-elasticity coefficient e2 is: μ agg(e2) =max{min(1 / 3, μ VS2(e2) ),min(1 / 3,μ S2(e2) ),min(1 / 3,μ S2(e2) ),min(1 / 3,μ M2(e2) ),min(1 / 3,μVS2(e2) ),min(1 / 3,μ S2(e2) ),min(1 / 2,μ S2(e2) ),min(1 / 2,μ S2(e2) )}=max{min(1 / 2, μ VS2(e2) ),min(1 / 2,μ S2(e2) ),min(1 / 3,μ M2(e2) The terminal can use the centroid method to perform defuzzification and obtain the elastic coefficients as follows: e1 = 0.0942, e2 = 0.0668. The self-elasticity coefficient e1 and mutual elasticity coefficient e2 of each object to be predicted are as follows:

[0106]

[0107] In the elasticity coefficient matrix, the self-elasticity coefficient is negative, and the mutual elasticity coefficient is positive. Based on the above calculation results, the terminal can define the self-elasticity coefficient e1 multiplied by -1 as the coefficient for the peak-to-peak period in the elasticity coefficient matrix. The coefficients for the flat-to-flat and valley-to-valley periods are defined as -2 / 3 of the self-elasticity coefficient e1. The mutual elasticity coefficient e2 is defined as the coefficient for the peak-to-valley period, and the peak-to-flat and valley-to-valley periods are defined as 1 / 2 of the mutual elasticity coefficient e2. The elasticity coefficient matrix for the metal products industry can then be as follows:

[0108]

[0109] The elasticity coefficient matrix of the textile industry can be shown as follows:

[0110]

[0111] The terminal can set the time-of-use electricity price scheme for the above-mentioned preset area to be 1.8:1:0.3 and 2:1:0.25, with the peak-valley price ratio being 6:1 and 8:1 respectively. Based on the proposed adjustable potential assessment model and the current time-of-use electricity price, the terminal obtains the load curves before and after the electricity price-based demand response for the metal products industry and the textile industry through modeling and simulation analysis. Figure 4 and Figure 5 As shown, Figure 4 Schematic diagram of the interface for the demand response potential prediction step in one embodiment. Figure 5 FIG. 1 is a schematic diagram of an interface for a demand response potential prediction step in an embodiment. Figure 4 and Figure 5 The chart includes various power consumption curves after DR (Demand Response). For example, Figure 4 The current DR electricity consumption curve 301, the electricity consumption curve 302 when the electricity price ratio is 6:1, and the electricity consumption curve 303 when the electricity price ratio is 8:1; Figure 5The current DR power consumption curve 401, the power consumption curve 402 when the power price ratio is 6:1, and the power consumption curve 403 when the power price ratio is 8:1. Figure 4 and Figure 5 It can be seen that when the electricity price of the object to be predicted changes, there will be changes in the electricity demand in response to the change in electricity price, and the demand response potential of the metal manufacturing industry is greater than that of the textile industry.

[0112] Through the above embodiment, the terminal reduces the prediction complexity by predicting the demand response potential based on the unit kilowatt-hour electricity output value, peak and valley load factor, load interruptibility factor and elasticity coefficient matrix.

[0113] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0114] Based on the same inventive concept, the embodiments of the present application also provide a demand response potential prediction device for implementing the above-mentioned demand response potential prediction method. The implementation solution provided by this device is similar to the implementation solution described in the above-mentioned method. Therefore, the specific limitations of one or more demand response potential prediction device embodiments provided below can be found in the above-mentioned limitations of the demand response potential prediction method and will not be repeated here.

[0115] In one embodiment, Figure 6 As shown, a demand response potential prediction device is provided, comprising: a response module 500, a query module 502, a determination module 505 and a prediction module 506, wherein:

[0116] Response module 500 is used to respond to a demand response potential prediction request, query and obtain the unit kilowatt-hour electricity output value, load peak-valley distribution coefficient and load interruptibility coefficient corresponding to the object to be predicted from the power database; the unit kilowatt-hour electricity output value represents the impact of electricity cost on the electricity consumption of the object to be predicted; the load peak-valley distribution coefficient represents the electricity consumption time distribution characteristics of the electricity consumption of the object to be predicted; the load interruptibility coefficient represents the degree of impact of interrupting electricity consumption on the production of the object to be predicted.

[0117] Query module 502 is used to query the preset rule table according to the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient to obtain the corresponding self-elasticity coefficient and mutual elasticity coefficient; the preset rule table includes the corresponding relationship between the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient and the self-elasticity coefficient and the mutual elasticity coefficient; the self-elasticity coefficient represents the corresponding relationship between the electricity consumption of the object to be predicted in each time period and the electricity cost of the current time period; the mutual elasticity coefficient represents the corresponding relationship between the electricity consumption of the object to be predicted in each time period and the electricity cost of other time periods.

[0118] The determination module 504 is used to determine the elasticity coefficient matrix based on the self-elasticity coefficient and the mutual elasticity coefficient, and determine the new electricity demand of the object to be predicted in the preset time period in response to the updated electricity cost based on the elasticity coefficient matrix, the original electricity demand of the object to be predicted in the preset time period, the original electricity cost of the object to be predicted in the preset time period, the updated electricity cost corresponding to the object to be predicted in the preset time period, and the number of electricity time periods.

[0119] The prediction module 506 is configured to determine the demand response potential of the object to be predicted based on the difference between the new power demand and the original power demand.

[0120] In one embodiment, the above-mentioned response module 500 is specifically used to obtain the annual output value increase and annual electricity consumption in the historical time period of the object to be predicted, and to obtain the total electricity consumption in the preset time period of the object to be predicted, the preset time period including the peak electricity consumption period, the valley electricity consumption period and the level electricity consumption period; determine the unit kilowatt-hour electricity output value of the object to be predicted based on the annual output value increase and the annual electricity consumption in the historical time period; obtain the first electricity consumption in the peak electricity consumption period, the second electricity consumption in the valley electricity consumption period, and the number of electricity consumption data points in the peak electricity consumption period, the valley electricity consumption period and the level electricity consumption period in the total electricity consumption in the preset time period, and obtain the load peak-valley distribution coefficient based on the first electricity consumption, the second electricity consumption and the number of electricity consumption data points; determine the load interruptibility coefficient corresponding to the object to be predicted based on the electricity consumption behavior of the object to be predicted.

[0121] In one embodiment, the above-mentioned device also includes: a division module, which is used to obtain the region where the object to be predicted is located, and determine the division strategy of the peak power consumption period, valley power consumption period and level power consumption period in the preset time period according to the electricity cost strategy corresponding to the region; determine the peak power consumption period, valley power consumption period and level power consumption period in the preset time period according to the division strategy.

[0122] In one embodiment, the above-mentioned division module is specifically used to determine the division strategy of peak electricity consumption period, valley electricity consumption period and level electricity consumption period in the preset time period according to the time-of-use electricity price strategy if the electricity cost strategy is a time-of-use electricity price strategy; if the electricity cost strategy is not a time-of-use electricity price strategy, determine the division strategy of peak electricity consumption period, valley electricity consumption period and level electricity consumption period in the preset time period according to the typical daily electricity consumption curve.

[0123] In one embodiment, the query module 502 is specifically used to input the unit kilowatt-hour electricity output value into a first membership function to obtain a plurality of first probabilities of a plurality of unit kilowatt-hour electricity output value levels corresponding to the unit kilowatt-hour electricity output value output by the first membership function; input the load peak-valley distribution coefficient into a second membership function to obtain a plurality of second probabilities of a plurality of load peak-valley distribution coefficient levels corresponding to the load peak-valley distribution coefficient output by the second membership function; input the load interruptibility coefficient into a third membership function to obtain a plurality of third probabilities of a plurality of load interruptibility coefficient levels corresponding to the load interruptibility coefficient output by the third membership function; and according to the plurality of unit kilowatt-hour electricity output value levels and the corresponding plurality of first probabilities, the plurality of load peak-valley distribution coefficient levels and the corresponding plurality of second probabilities, and the plurality of load interruptibility coefficient levels and the corresponding Multiple third probabilities query a preset rule table to obtain multiple fourth probabilities of multiple self-elasticity coefficient levels and multiple fifth probabilities of multiple mutual elasticity coefficient levels under the conditions of each group of unit kilowatt-hour power output value level, load peak-valley distribution coefficient level and load interruptibility coefficient level; perform logical AND operations on the multiple fourth probabilities and the multiple fifth probabilities respectively to determine at least one target self-elasticity coefficient level corresponding to at least one target fourth probability with the largest probability corresponding to the object to be predicted, and determine at least one target mutual elasticity coefficient level corresponding to at least one target fifth probability with the largest probability corresponding to the object to be predicted; defuzzify the at least one target self-elasticity coefficient level and the at least one target mutual elasticity coefficient level according to the center of gravity method to obtain the self-elasticity coefficient and mutual elasticity coefficient of the object to be predicted.

[0124] In one embodiment, the above-mentioned determination module 504 is specifically used to determine, based on the self-elasticity coefficient, a first coefficient between the electricity consumption valley period and the electricity consumption valley period, a second coefficient between the electricity consumption level period and the electricity consumption level period, and a third coefficient between the electricity consumption peak period and the electricity consumption peak period within a preset time period; based on the mutual elasticity coefficient, determine, based on the mutual elasticity coefficient, a fourth coefficient between the electricity consumption peak period and the electricity consumption valley period, a fifth coefficient between the electricity consumption peak period and the electricity consumption level period, and a sixth coefficient between the electricity consumption level period and the electricity consumption valley period within the preset time period; obtain an elasticity coefficient matrix based on the first coefficient, the second coefficient, the third coefficient, the fourth coefficient, the fifth coefficient, and the sixth coefficient; obtain the product of the elasticity coefficient matrix, the original electricity demand, the electricity cost change obtained by the difference between the original electricity cost and the updated electricity cost, and the ratio of the original electricity cost, and determine, based on the ratio of the product to the number of electricity consumption periods, a new electricity load change of the object to be predicted in response to the updated electricity cost within the preset time period, and obtain the updated electricity demand based on the sum of the new electricity load change and the original electricity demand.

[0125] Each module in the aforementioned demand response potential prediction device may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a computer device memory in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0126] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a demand response potential prediction method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse.

[0127] Those skilled in the art will understand that Figure 7The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0128] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the above-mentioned demand response potential prediction method when executing the computer program.

[0129] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned demand response potential prediction method is implemented.

[0130] In one embodiment, a computer program product is provided, comprising a computer program, which implements the above-mentioned demand response potential prediction method when executed by a processor.

[0131] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0132] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0133] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0134] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A demand response potential prediction method, characterized in that: The method comprises: In response to a demand response potential prediction request, query and obtain the unit kilowatt-hour electricity output value, load peak-valley distribution coefficient, and load interruptibility coefficient corresponding to the object to be predicted from the power database; the unit kilowatt-hour electricity output value represents the impact of electricity cost on the electricity consumption of the object to be predicted; the load peak-valley distribution coefficient represents the distribution characteristics of the electricity consumption of the object to be predicted during the electricity consumption period; and the load interruptibility coefficient represents the degree of impact of interrupting electricity consumption on the production of the object to be predicted; According to the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient, a preset rule table is queried to obtain the corresponding self-elasticity coefficient and mutual elasticity coefficient; the preset rule table includes the corresponding relationship between the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient and the self-elasticity coefficient and the mutual elasticity coefficient; the self-elasticity coefficient represents the corresponding relationship between the electricity consumption of the object to be predicted in each time period and the change in electricity cost in the current time period; the mutual elasticity coefficient represents the corresponding relationship between the electricity consumption of the object to be predicted in each time period and the change in electricity cost in other time periods; determining an elasticity coefficient matrix according to the self-elasticity coefficient and the mutual elasticity coefficient, and determining a new electricity demand of the object to be predicted in the preset time period in response to the updated electricity cost according to the elasticity coefficient matrix, the original electricity demand of the object to be predicted in the preset time period, the original electricity cost of the object to be predicted in the preset time period, the updated electricity cost corresponding to the object to be predicted in the preset time period, and the number of electricity time periods; The demand response potential of the object to be predicted is determined according to the difference between the new electricity demand and the original electricity demand.

2. The method according to claim 1, characterized in that The obtaining of the unit kilowatt-hour electricity output value, the load peak-valley distribution coefficient, and the load interruptibility coefficient corresponding to the object to be predicted includes: Obtaining the annual output value increase and annual electricity consumption of the object to be predicted within a historical time period, and obtaining the total electricity consumption of the object to be predicted within a preset time period, wherein the preset time period includes a peak electricity consumption period, a valley electricity consumption period, and a level electricity consumption period; Determine the unit kilowatt-hour output value of the object to be predicted based on the annual output value increase and annual electricity consumption in the historical time period; Obtaining a first power consumption during a peak period, a second power consumption during a valley period, and the number of power consumption data points during the peak period, the valley period, and the level period from the total power consumption during the preset time period, and obtaining a load peak-valley distribution coefficient based on the first power consumption, the second power consumption, and the number of power consumption data points; A load interruptibility coefficient corresponding to the object to be predicted is determined according to the power consumption behavior of the object to be predicted.

3. The method according to claim 2, characterized in that Before obtaining the first power consumption during the peak period, the second power consumption during the valley period, and the number of power consumption data points during the peak period, the valley period, and the level period, of the total power consumption during the preset time period, the method further includes: Obtaining the region where the object to be predicted is located, and determining a division strategy for the preset time period into peak power consumption period, valley power consumption period, and level power consumption period according to the power consumption cost strategy corresponding to the region; The peak power consumption period, the valley power consumption period and the level power consumption period within the preset time period are determined according to the division strategy.

4. The method according to claim 3, characterized in that The determining, based on the electricity cost strategy corresponding to the region, a division strategy of the preset time period into a peak electricity consumption period, a valley electricity consumption period, a level electricity consumption period, and a level electricity consumption period, includes: If the electricity cost strategy is a time-of-use electricity price strategy, a strategy for dividing the preset time period into a peak electricity consumption period, a valley electricity consumption period, and a level electricity consumption period is determined according to the time-of-use electricity price strategy; If the electricity cost strategy is not a time-of-use electricity price strategy, a division strategy of the preset time period into peak electricity consumption period, valley electricity consumption period and level electricity consumption period is determined according to a typical daily electricity consumption curve.

5. The method according to claim 1, characterized in that The step of querying a preset rule table according to the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient to obtain the corresponding self-elasticity coefficient and mutual elasticity coefficient includes: Inputting the unit kilowatt-hour electricity output value into a first membership function to obtain a plurality of first probabilities of a plurality of unit kilowatt-hour electricity output value levels corresponding to the unit kilowatt-hour electricity output value output by the first membership function; Inputting the load peak-valley distribution coefficient into a second membership function to obtain a plurality of second probabilities of a plurality of load peak-valley distribution coefficient levels corresponding to the load peak-valley distribution coefficient output by the second membership function; Inputting the load interruptibility coefficient into a third membership function to obtain a plurality of third probabilities of a plurality of load interruptibility coefficient levels corresponding to the load interruptibility coefficient output by the third membership function; According to the multiple unit kilowatt-hour electricity production value levels and the corresponding multiple first probabilities, the multiple load peak-valley distribution coefficient levels and the corresponding multiple second probabilities, and the multiple load interruptibility coefficient levels and the corresponding multiple third probabilities, a preset rule table is searched to obtain multiple fourth probabilities of multiple self-elasticity coefficient levels and multiple fifth probabilities of multiple mutual elasticity coefficient levels under the conditions of each group of unit kilowatt-hour electricity production value levels, load peak-valley distribution coefficient levels, and load interruptibility coefficient levels; performing a logical AND operation on the plurality of fourth probabilities and the plurality of fifth probabilities, respectively, to determine at least one target self-elasticity coefficient level corresponding to the fourth probability of at least one target having the greatest probability corresponding to the object to be predicted, and to determine at least one target mutual elasticity coefficient level corresponding to the fifth probability of at least one target having the greatest probability corresponding to the object to be predicted; Defuzzification is performed on the at least one target self-elasticity coefficient level and the at least one target mutual-elasticity coefficient level respectively according to the center of gravity method to obtain the self-elasticity coefficient and the mutual-elasticity coefficient of the object to be predicted.

6. The method according to claim 1, characterized in that The determining of an elasticity coefficient matrix according to the self-elasticity coefficient and the mutual elasticity coefficient, and determining a new electricity demand of the object to be predicted in the preset time period in response to the updated electricity cost according to the elasticity coefficient matrix, the original electricity demand of the object to be predicted in the preset time period, the original electricity cost of the object to be predicted in the preset time period, the updated electricity cost corresponding to the object to be predicted in the preset time period, and the number of electricity time periods, includes: Determine, based on the self-elasticity coefficient, a first coefficient between valley periods, a second coefficient between level periods, and a third coefficient between peak periods within the preset time period; Determining, based on the mutual elasticity coefficient, a fourth coefficient between the peak power consumption period and the valley power consumption period, a fifth coefficient between the peak power consumption period and the level power consumption period, and a sixth coefficient between the level power consumption period and the valley power consumption period within the preset time period; Obtaining the elastic coefficient matrix according to the first coefficient, the second coefficient, the third coefficient, the fourth coefficient, the fifth coefficient and the sixth coefficient; Obtain the product of the elasticity coefficient matrix, the original electricity demand, the electricity cost change obtained by the difference between the original electricity cost and the updated electricity cost, and the ratio of the original electricity cost, and determine the new electricity load change of the object to be predicted in response to the updated electricity cost within a preset time period based on the ratio of the product to the number of electricity time periods, and obtain the updated electricity demand based on the sum of the new electricity load change and the original electricity demand.

7. A demand response potential prediction device, characterized in that: The device comprises: A response module is configured to query and obtain, from a power database, a unit kilowatt-hour electricity output value, a load peak-valley distribution coefficient, and a load interruptibility coefficient corresponding to an object to be predicted in response to a demand response potential prediction request; the unit kilowatt-hour electricity output value represents the impact of electricity cost on the electricity consumption of the object to be predicted; the load peak-valley distribution coefficient represents the distribution characteristics of the electricity consumption of the object to be predicted during the electricity consumption period; and the load interruptibility coefficient represents the degree of impact of interrupting electricity consumption on the production of the object to be predicted; A query module is configured to query a preset rule table based on the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient to obtain corresponding self-elasticity coefficients and mutual elasticity coefficients; the preset rule table includes a correspondence between the level corresponding to the unit kilowatt-hour electricity output value, the level corresponding to the load peak-valley distribution coefficient, and the level corresponding to the load interruptibility coefficient and the self-elasticity coefficient and the mutual elasticity coefficient; the self-elasticity coefficient represents the correspondence between the electricity consumption of the object to be predicted in each time period and the change in electricity cost in the current time period; the mutual elasticity coefficient represents the correspondence between the electricity consumption of the object to be predicted in each time period and the change in electricity cost in other time periods; a determination module, configured to determine an elasticity coefficient matrix based on the self-elasticity coefficient and the mutual elasticity coefficient, and determine a new electricity demand of the object to be predicted within the preset time period in response to the updated electricity cost based on the elasticity coefficient matrix, the original electricity demand of the object to be predicted within the preset time period, the original electricity cost of the object to be predicted within the preset time period, the updated electricity cost corresponding to the object to be predicted within the preset time period, and the number of electricity time periods; The prediction module is used to determine the demand response potential of the object to be predicted according to the difference between the new electricity demand and the original electricity demand.

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

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

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

Images