A method and system for guiding demand side participation in flexible interaction considering the influence of social information, and a storage medium

Abstract

The application discloses a kind of guiding method, system and storage medium of demand side participation flexible interaction considering social information influence.Process is: the influence degree of each social information is determined to user power load respectively;According to the social information corresponding to the influence degree, the user power consumption is determined;Establish the multi-level market electricity purchase model of power seller and the optimization decision model of power purchase and sale of power seller;The optimal demand side adjustable load is obtained by solving the optimization decision model of power purchase and sale of power seller, and the user is guided to participate in flexible interaction according to the optimal demand side adjustable load.The application can accurately calculate the relationship between guiding information and user load adjustment amount, and obtain the optimal guiding strategy by combining optimization algorithm.The obtained guiding strategy can effectively achieve the optimization goal, and also play a certain degree of peak shaving and valley filling role, solve the problem of power load peak rise damaging power grid stability, have certain social benefits, and have strong applicability.

Description

Technical Field

[0001] This invention belongs to the field of flexible interaction guidance technology on the demand side, specifically relating to a guidance method, system, and storage medium for flexible interaction of demand side participation that takes into account the impact of social information. Background Technology

[0002] With the rapid development of society, economy, and technology, especially the widespread adoption of smart devices such as smart home appliances, electricity demand has increased significantly and possesses enormous adjustable potential. Electricity retailers can regulate this adjustable load by releasing reasonable and effective guidance information, encouraging users to participate in flexible interactions. "Guidance information" refers to a series of messages released by electricity retailers on the internet aimed at guiding users to use electricity rationally and actively participate in flexible interactions. Methods such as releasing time-of-use pricing, energy discount coupons, social activity coupons, and energy-saving information pushes, which incentivize users to use electricity rationally, are receiving increasing attention.

[0003] However, current research on flexible interactive guidance methods for the demand side assumes users participate as absolutely rational individuals. But in reality, users' electricity consumption behavior is not entirely rational. The rationality of their electricity decisions is limited by factors such as known information and cognitive limitations, exhibiting typical characteristics of bounded rationality. Furthermore, in electricity consumption behavior, users are satisfied rather than maximizers, meaning that traditional guidance methods may not necessarily achieve the goal of guiding users to adjust their electricity consumption. Moreover, existing guidance methods do not consider the impact of social information such as temperature, humidity, and electricity prices on users' electricity load. The accuracy of load changes under the guidance information is low, and in real life, users may violate the theoretically optimal electricity adjustment amount, causing the guidance information to fail. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of the aforementioned background technology by providing a guidance method, system, and storage medium that considers the impact of social information on demand-side participation in flexible interaction, thereby solving the problem of peak power load disrupting grid stability.

[0005] The technical solution adopted in this invention is: a guidance method for flexible interaction of demand-side participation that takes into account the impact of social information.

[0006] Determine the degree of impact of each piece of social information on users' electricity load;

[0007] The user's electricity consumption is determined based on the social information corresponding to the degree of impact.

[0008] Establish a multi-level market electricity purchase model for e-commerce retailers;

[0009] An optimized decision-making model for electricity purchase and sales by electricity retailers is established based on social information, user electricity consumption, and a multi-level market electricity purchase model of electricity retailers.

[0010] The optimal demand-side adjustable load is obtained by solving the electricity purchase and sale optimization decision model of the e-commerce platform using the particle swarm optimization algorithm and the CPLEX solver. Based on the optimal demand-side adjustable load, users are guided to participate in flexible interaction.

[0011] Furthermore, the impact of social information on user electricity load is quantified using a maximum information coefficient model. This social information includes coupon coefficients, historical user electricity load, temperature, and humidity. The maximum information coefficient model is...

[0012]

[0013] In the formula: MIC represents the correlation between social information and user electricity load; X represents social information; Y represents user electricity load; n x ,n y G represents the number of grid cells along the x and y axes; G is n. x ×n y The formed grid; I G (X,Y) represents the mutual information under grid G; B(n,α)=n α (0 < α < 1) is a function used to limit the maximum number of cells; log2 min(n x ,n y ) is a standardization item that ensures that MIC is in the range of 0 to 1.

[0014] Furthermore, the coupon coefficient is determined using the following formula:

[0015]

[0016] In the formula: cp i,t Let k be the coupon amount that user i can obtain at time t; t Let Δq be the coupon coefficient at time t; i,t Let i be the load adjustment amount for user i at time t; The load adjustment amount required for user i to obtain the maximum coupon amount at time t; The maximum load adjustment for user i at time t; Let be the maximum coupon amount that user i can obtain at time t.

[0017] Furthermore, the user's electricity consumption is calculated using the Attention-LSTM load prediction model:

[0018]

[0019] x t =[Te j (t), hj (t), p j (t), k j (t), q j (t-1), q j (t-2), q j-1 (t), q j-1 (t-1)]

[0020] In the formula: f represents the user's electricity consumption at time t; t Here is the electricity load forecasting function; x t Let be the input feature vector at time t; j represents the number of days; Te j (t) represents the predicted temperature at the predicted time, h j (t) represents the predicted humidity and p at the predicted time. j (t) represents the electricity price at the predicted time, k j (t) represents the coupon coefficient at the predicted time point, q j (t-1) represents the load value at the time before the prediction point, q j (t-2) represents the load value two moments before the prediction point, q j-1 (t) represents the load value at the same time the day before the prediction point, q j-1 (t-1) represents the load value at the time one day before the prediction point.

[0021] Furthermore, the multi-tiered market power purchase model of the electricity retailer includes a medium- and long-term market power purchase model and a day-ahead market power purchase model, respectively.

[0022]

[0023] In the formula: B Y For the cost of purchasing electricity in the medium to long term market; B D Y represents the cost of electricity purchase in the day-ahead market; D represents the cost in the medium- and long-term market; L represents the number of medium- and long-term contracts signed by electricity retailers. This refers to the electricity volume stipulated in medium- and long-term contracts. The price of a medium- to long-term contract; The amount of electricity purchased at time t in the market before the current date; The electricity price at time t in the market is as follows.

[0024] Furthermore, the online retailer's electricity purchase and sales optimization decision model is as follows:

[0025] max C = (C sell -C buy -C p -C cp )

[0026]

[0027] C buy =B Y +B D

[0028]

[0029]

[0030]

[0031] Where: C represents the total revenue of the e-commerce platform; C sell For electricity sales revenue; C buy For electricity purchase expenses; C P C. Settle electricity charges for deviations in electricity volume paid by the electricity sales company to the power generation company; cp I represents the cost of coupons; I represents the user set; p represents the cost of coupons. t Let t be the peak-valley time-of-use electricity price at time t; T is the set of times throughout the day; B represents the electricity consumption of user i at time t. Y For the cost of purchasing electricity in the medium to long term market; B D The current daytime electricity purchase cost; The deviation of the bilateral contract for electricity sales company at time t; and These represent the settlement prices for positive and negative deviations in electricity sales by the electricity sales company; α1 and α2 are 0 / 1 variables. When the deviation in electricity sales by the electricity sales company is positive, α1 = 1 and α2 = 0, otherwise α1 = 0 and α2 = 1. The electricity price at time t in the market before the current date; Let be the proportion of the electrical energy allocated to contract l at time t.

[0032] Furthermore, the process of solving the optimal decision-making model for electricity sales by e-commerce platforms includes the following steps:

[0033] Step 1: Input the parameters required for the e-commerce e-commerce purchase and sales optimization decision model;

[0034] Step 2: Set the number of particles and the maximum number of iterations, where each particle represents a coupon coefficient;

[0035] Step 3: Randomly generate an initial particle swarm, calculate the fitness value of all individuals in the initial particle swarm, and obtain the individual extreme values ​​and the global extreme value.

[0036] Step 4: Each particle follows two extreme values ​​to change its velocity and position, and then compares them with the individual extreme value and the global extreme value. If it is better than the individual extreme value or the global extreme value, the individual extreme value or the global extreme value is updated, and the position and velocity of each particle are updated with the updated individual extreme value and the global extreme value to form a new particle.

[0037] Step 5: Call the user electricity consumption behavior analysis program, and use Attention-LSTM to solve the electricity consumption of each user at each time based on the updated particles;

[0038] Step 6: Call the electricity sales e-commerce optimization subroutine, and use the CPLEX solver to solve for the electricity sales e-commerce revenue based on the electricity consumption of each user at each time, with the objective function of maximizing benefits.

[0039] Step 7: Increment the current iteration count by one and compare it with the set maximum iteration count. If the maximum iteration count has not been reached, return to step 4; otherwise, end the iteration process and determine the optimal demand-side adjustable load using the particle updated in the last iteration count.

[0040] Furthermore, the optimal demand-side adjustable load is determined based on the user's electricity consumption at a corresponding time of day, using the optimal coupon coefficient, and the user is guided to participate in load regulation based on the electricity consumption.

[0041] A system for implementing a guidance method that considers the impact of social information on demand-side participation and flexible interaction, as described in any of the above embodiments, comprising:

[0042] The data acquisition module is used to collect social information data;

[0043] The user electricity consumption prediction module determines the degree of impact of each piece of social information on the user's electricity load, and determines the user's electricity consumption based on the social information corresponding to the degree of impact.

[0044] The model building module is used to build a multi-level market electricity purchase model for electricity retailers, and to build an optimized decision-making model for electricity purchase and sales by electricity retailers based on social information, user electricity consumption, and the multi-level market electricity purchase model of electricity retailers.

[0045] The flexible interaction module is used to solve the electricity purchase and sale optimization decision model of the e-commerce platform using the particle swarm optimization algorithm and the CPLEX solver to obtain the optimal demand-side adjustable load, and guide users to participate in flexible interaction based on the optimal demand-side adjustable load.

[0046] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in any of the preceding claims.

[0047] The beneficial effects of this invention are as follows:

[0048] (1) This invention considers the impact of social information and predicts the user's electricity load after the effect of social information, so that the electricity retailers can formulate appropriate guidance strategies to guide users to complete the load interaction target more accurately based on the target amount of each user load interaction; based on social information, user electricity consumption, and the electricity purchase model of the multi-level market of the electricity retailers, an optimal decision-making model for electricity purchase and sale of the electricity retailers is established, which can obtain the optimal load amount for user participation in the interaction, thereby formulating appropriate guidance strategies to guide users to complete the load interaction target more accurately; and based on the established model, the electricity purchase volume of the electricity retailers at each level of the market can be reasonably allocated, effectively reducing the electricity purchase and sale risk of the electricity retailers, while effectively guiding users to reduce electricity consumption at peak times and increase electricity consumption at valley times, reducing the user's electricity cost and having certain economic benefits.

[0049] (2) The guidance method proposed in this invention, which considers the impact of social information on the demand side and participates in flexible interaction, can accurately calculate the relationship between guidance information and user electricity consumption, and obtain the optimal guidance strategy by combining optimization algorithm. The obtained guidance strategy can effectively achieve the optimization goal, and at the same time play a certain role in peak shaving and valley filling, solving the problem of power load peak rise damaging the stability of the power grid. It has certain social benefits and strong applicability. Attached Figure Description

[0050] Figure 1 This is the control flowchart of the present invention.

[0051] Figure 2 This is the MIC diagram of social information and user electricity load in this invention.

[0052] Figure 3 This is a structural diagram of the Attention-LSTM load prediction model of the present invention.

[0053] Figure 4 This is a graph showing how the coupon coefficient of this invention changes with the user load adjustment amount.

[0054] Figure 5 E-commerce sales structure diagram.

[0055] Figure 6 This is a flowchart illustrating the solution of the online e-commerce purchase and sales optimization decision model of this invention.

[0056] Figure 7 This is a user load curve before and after the implementation of the guidance method of the present invention. Detailed Implementation

[0057] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings. It should be noted that these descriptions are for the purpose of aiding understanding the present invention, but do not constitute a limitation thereof. Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0058] like Figure 1-6 As shown, this invention provides a method for guiding flexible interaction on the demand side that considers the impact of social information. The process is as follows:

[0059] Determine the degree of impact of each piece of social information on users' electricity load;

[0060] Determine user electricity consumption based on social information corresponding to the degree of impact;

[0061] Establish a multi-level market electricity purchase model for e-commerce retailers;

[0062] An optimized decision-making model for electricity purchase and sales by electricity retailers is established based on social information, user electricity consumption, and a multi-level market electricity purchase model of electricity retailers.

[0063] The optimal demand-side adjustable load is obtained by solving the electricity purchase and sale optimization decision model of the e-commerce platform using the particle swarm optimization algorithm and the CPLEX solver. Based on the optimal demand-side adjustable load, users are guided to participate in flexible interaction.

[0064] In the above scheme, the maximum information coefficient (MIC) can be plotted on a scatter plot of two variables to encapsulate the correlation between them, and is unaffected by the distribution of the variable pair and the type of correlation. Therefore, this invention quantifies the impact of social information on user electricity load through the maximum information coefficient model. This social information includes coupon coefficients, historical user electricity load (behavioral habits), temperature, and humidity. The maximum information coefficient model is as follows:

[0065]

[0066] In the formula: MIC represents the correlation between social information and user electricity load; X represents social information, i.e., X = (x1, x2, ..., x...). n Each element in the vector is an n-dimensional vector, and each component of the vector is a real number. Y represents the user's electricity load. x ,n y G represents the number of grid cells along the x and y axes; G is n. x ×n y The formed grid; I G (X,Y) represents the mutual information under grid G; B(n,α)=n α (0 < α < 1) is a function used to limit the maximum number of cells, where n is the amount of data; log2 min(n x ,n y ) is a standardization term that ensures that MIC is in the range of 0 to 1. The larger the MIC, the greater the corresponding impact.

[0067] In the above scheme, since coupons need to take into account the different load adjustment characteristics of different users—that is, under the same coupon incentive, different users will have different load adjustment amounts—the coupon amount should be positively correlated with the user's load adjustment amount. Considering fairness, different users should have the same coupon coefficient; the coupon amount should be adjusted within a reasonable range, avoiding extreme values. A piecewise linear function is used to represent the relationship between the coupon amount and the user's load adjustment amount. The decrease in load during peak hours is positive, and the increase in load during off-peak hours is positive. The coupon coefficient is determined by the following formula:

[0068]

[0069] In the formula: cp i,t Let k be the coupon amount that user i can obtain at time t; t Let Δq be the coupon coefficient at time t; i,t Let i be the load adjustment amount for user i at time t; The load adjustment amount required for user i to obtain the maximum coupon amount at time t; The maximum load adjustment for user i at time t; Let be the maximum coupon amount that user i can obtain at time t.

[0070] When the user load adjustment increases positively, the coupon value also increases positively, thus allowing users who respond actively to each demand to receive greater discounts. Simultaneously, different coupon coefficients k... t The value of the coupon coefficient varies, and the rate at which the coupon amount changes with the load adjustment varies. The larger the coupon coefficient, the larger the coupon the user receives per unit of adjustment, but the coupon amount can remain stable between 0 and 1. Between, therefore regardless of k t How to determine the value ensures that the user's electricity purchase cost can always be kept within a reasonable range.

[0071] In the above scheme, since different social information has varying degrees of impact on user electricity consumption, an Attention mechanism is introduced. Social information with a high degree of impact (greater than 0.01) is selected as input features. By assigning different weights to the input features of the model, more critical impact information is highlighted. The user electricity consumption is then calculated using the Attention-LSTM load prediction model.

[0072]

[0073] x t =[Te j (t), h j (t), pj (t), k j (t), q j (t-1), q j (t-2), q j-1 (t), q j-1 (t-1)]

[0074] In the formula: f represents the user's electricity consumption at time t; t Here is the electricity load forecasting function; x t Let be the input feature vector at time t; j represents the number of days; Te j (t) represents the predicted temperature at the predicted time, h j (t) represents the predicted humidity and p at the predicted time. j (t) represents the electricity price at the predicted time, k j (t) represents the coupon coefficient at the predicted time point, q j (t-1) represents the load value at the time before the prediction point, q j (t-2) represents the load value two moments before the prediction point, q j-1 (t) represents the load value at the same time the day before the prediction point, q j-1 (t-1) represents the load value at the time one day before the prediction point.

[0075] The Attention-LSTM load prediction model consists of two parts: the first part is a two-layer LSTM, where the aforementioned social information is input into the first layer Pre-LSTM for pre-training, and the output is h. t With state s t The output of the first layer is fed into the second LSTM model for training. The second part is the Attention layer, which assigns the feature weights learned by the model to the input vector in the next time step, highlighting the impact of key features on the predicted load. Finally, the data passes through a fully connected layer, is adjusted to the specified vector format, and then the final output layer is the user's electricity load at the predicted time.

[0076] In the above scheme, the multi-level market power purchase model of the electricity retailer includes a medium- and long-term market power purchase model and a day-ahead market power purchase model.

[0077] 1. Establish a medium- to long-term market-based electricity purchase model;

[0078] Electricity retailers purchase power through medium- to long-term bilateral physical contracts with power generation companies. These contracts are segmented contracts, which divide the load into different durations based on the continuous production characteristics of electricity. Each duration is governed by a different contract. The formula for calculating the electricity purchase cost of electricity retailers under these medium- to long-term bilateral physical contracts is as follows:

[0079]

[0080] In the formula: B Y Y represents the cost of purchasing electricity in the medium- and long-term market; L represents the medium- and long-term market; and L represents the number of medium- and long-term contracts signed by the electricity retailers. This refers to the electricity volume stipulated in medium- and long-term contracts. The price is for medium- to long-term contracts.

[0081] 2. Establish a day-ahead market electricity purchase model;

[0082] Because there is a discrepancy between the contracted electricity volume in the medium- and long-term contracts signed by electricity retailers and the actual electricity consumption of users, in order to reduce the penalty for this discrepancy, electricity retailers need to conduct day-ahead short-term load forecasting. Then, based on the discrepancy between the contracted electricity volume and the forecasted electricity volume, they purchase electricity in the day-ahead market to reduce the penalty for this discrepancy and lower the electricity purchase risk. The formula for calculating the cost of electricity purchased by electricity retailers in the day-ahead market is as follows:

[0083]

[0084] In the formula: B D D represents the day-ahead market electricity purchase cost; T represents the day-ahead market price; and T represents the total time of day. The amount of electricity purchased at time t in the market before the current date; The electricity price at time t in the market is as follows.

[0085] In the above scheme, the electricity retailer aims to maximize its electricity purchase and sale revenue. The retailer's revenue is influenced by both actual user electricity consumption and purchase costs. Electricity sales revenue is determined by user load, while purchase costs are determined by the purchase method and quantity. Therefore, this invention uses coupons to guide users to change their electricity consumption behavior, thereby optimizing the electricity purchase strategy and maximizing the retailer's electricity purchase and sale revenue. The objective function expression of the electricity retailer's electricity purchase and sale optimization decision model is as follows:

[0086] max C = (C sell -C buy -C p -C cp )

[0087] In the formula: C represents the total revenue of the e-commerce platform; C sell For electricity sales revenue; C buy For electricity purchase expenses; C P C. Settle electricity charges for deviations in electricity volume paid by the electricity sales company to the power generation company; cp Cost of coupons;

[0088] Electricity sales revenue C sell The expression is as follows:

[0089]

[0090] In the formula: I represents the user set; p t Let t be the peak-valley time-of-use electricity price; T is the set of times throughout the day; Let represent the electricity consumption of user i at time t.

[0091] Electricity purchase expenditure C buy The expression is as follows:

[0092] C buy =B Y +B D

[0093] The electricity sales company pays the power generation company the deviation electricity amount settlement fee C P The expression is as follows:

[0094]

[0095]

[0096] In the formula: The bilateral contract deviation of the electricity sales company at time t; and α1 and α2 are the settlement prices for positive and negative deviations in electricity consumption for the electricity sales company, respectively; α1 and α2 are 0 / 1 variables, where α1 = 1 and α2 = 0 when the deviation is positive, and α1 = 0 and α2 = 1 when the deviation is negative; ω l,t Let l be the proportion of the electrical energy allocated to contract l at time t.

[0097] Coupon cost C cp The expression is as follows:

[0098]

[0099] In the above scheme, a particle swarm optimization algorithm and a CPLEX solver are used to solve the electricity purchase and sale optimization decision model for the electricity retailers. For users, given a coupon coefficient, an Attention-LSTM model is used to predict user electricity load and feed it back to the electricity retailers' purchase and sale decision model. For the electricity retailers, both the particle swarm optimization algorithm, using the retailers' interests as the fitness function, is used to solve for the optimal coupon coefficient during the iteration process, and the CPLEX algorithm is used to solve for the optimal purchase volume for each market based on user electricity load. The optimal coupon coefficient guides users to participate in load regulation. The specific steps are as follows:

[0100] Step 1: Input the parameters required for the electricity purchase and sale optimization decision model of the electricity retailer, specifically including time-of-use pricing, unit electricity price of segmented bilateral contracts, and the electricity allocation ratio of the contract;

[0101] Step 2: Set the number of particles and the maximum number of iterations, where the particles are coupon coefficients;

[0102] Step 3: Randomly generate an initial particle swarm, calculate the fitness value of all individuals in the initial particle swarm, and obtain the individual extreme values ​​and the global extreme value.

[0103] Step 4: Each particle follows two "extreme values" to change its velocity and position, and then compares them with the individual extreme value and the global extreme value. If it is better than the individual extreme value or the global extreme value, the individual extreme value or the global extreme value is updated, and the position and velocity of each particle are updated with the updated individual extreme value and the global extreme value to form a new particle.

[0104] Step 5: Call the user electricity consumption behavior analysis program (i.e., the user electricity consumption calculation method mentioned above), and solve the electricity consumption of each user at each time based on Attention-LSTM;

[0105] Step 6: Call the electricity sales optimization subroutine (i.e., the electricity sales purchase and sale optimization decision model mentioned above), and solve for the electricity sales revenue of the electricity sales through the CPLEX solver with the objective function of maximizing benefits.

[0106] Step 7: Increment the current iteration count by one and compare it with the set maximum iteration count. If the maximum iteration count has not been reached, return to step 4; otherwise, end the optimization process and determine the optimal demand-side adjustable load using the particles updated in the last iteration count.

[0107] Step 8: The optimal demand-side adjustable load is the electricity consumption of a user at a corresponding time of day, determined based on the optimal coupon coefficient, and the user is guided to participate in load regulation based on the electricity consumption.

[0108] To achieve the aforementioned guidance method for flexible interaction of demand-side participation that considers the impact of social information, this invention also provides a guidance system for flexible interaction of demand-side participation that considers the impact of social information, including...

[0109] The data acquisition module is used to collect social information data;

[0110] The user electricity consumption prediction module determines the degree of impact of each piece of social information on the user's electricity load, and determines the user's electricity consumption based on the social information corresponding to the degree of impact.

[0111] The model building module is used to build a multi-level market electricity purchase model for electricity retailers, and to build an optimized decision-making model for electricity purchase and sales by electricity retailers based on social information, user electricity consumption, and the multi-level market electricity purchase model of electricity retailers.

[0112] The flexible interaction module is used to solve the electricity purchase and sale optimization decision model of the e-commerce platform using the particle swarm optimization algorithm and the CPLEX solver to obtain the optimal demand-side adjustable load, and guide users to participate in flexible interaction based on the optimal demand-side adjustable load.

[0113] Using real load data from Australia in 2006 for simulation analysis, the proposed guidance method that considers flexible interaction and demand-side participation based on social information is applied to regulate user load. User load before and after the strategy implementation is as follows: Figure 7 As shown in the figure, the results demonstrate that the proposed guidance method effectively guides users to use electricity rationally, reducing electricity consumption during peak hours and increasing it during off-peak hours. This not only reduces the electricity purchase costs for retailers but also achieves peak shaving and valley filling effects, thereby improving grid security.

[0114] In summary, the guidance method for flexible interaction of demand side participation that takes into account the impact of social information provided by this invention is effective and reasonable.

[0115] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described booting method and embodiments. Alternatively, when executed by a processor, the computer program implements the functions of each module / unit in the above-described booting system.

[0116] Those skilled in the art will also understand that the various illustrative logical blocks, units, and steps listed in the embodiments of the present invention can be implemented by electronic hardware, computer software, or a combination of both. To clearly demonstrate the interchangeability of hardware and software, the functions of the various illustrative components, units, and steps described above have been generally described. Whether such functionality is implemented through hardware or software depends on the specific application and the overall system design requirements. Those skilled in the art can implement the described functions using various methods for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of the present invention.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the scope defined by the claims of the present invention. Content not described in detail in this specification belongs to the prior art known to those skilled in the art.

Claims

1. A guidance and control method for flexible interaction of demand-side participation considering the impact of social information, characterized in that: Determine the degree of impact of each piece of social information on the user's electricity load. The social information includes coupon coefficients, the user's historical electricity load, temperature, and humidity. The user's electricity consumption is determined based on the social information corresponding to the degree of impact. Establish a multi-level market electricity purchase model for e-commerce retailers; An optimized decision-making model for electricity purchase and sales by electricity retailers is established based on social information, user electricity consumption, and a multi-level market electricity purchase model of electricity retailers. The optimal demand-side adjustable load is obtained by solving the electricity purchase and sale optimization decision model of the e-commerce platform using the particle swarm optimization algorithm and CPLEX solver. Based on the optimal demand-side adjustable load, users are guided to participate in flexible interaction. The coupon coefficient is determined by the following formula: In the formula: cp i,t Let k be the coupon amount that user i can obtain at time t; t Let Δq be the coupon coefficient at time t; i,t Let i be the load adjustment amount for user i at time t; The load adjustment amount required for user i to obtain the maximum coupon amount at time t; The maximum load adjustment for user i at time t; Let be the maximum coupon amount that user i can obtain at time t. The electricity sales and purchase optimization decision-making model is as follows: max C=(C sell -C buy -C p -C cp ) C buy =B Y +B D Where: C represents the total revenue of the e-commerce platform; C sell For electricity sales revenue; C buy For electricity purchase expenses; C P C. Settle electricity charges for deviations in electricity volume paid by the electricity sales company to the power generation company; cp I represents the cost of coupons; I represents the user set; p represents the cost of coupons. t Let t be the peak-valley time-of-use electricity price at time t; T is the set of times throughout the day; B represents the electricity consumption of user i at time t. Y For the cost of purchasing electricity in the medium to long term market; B D The current daytime electricity purchase cost; The deviation of the bilateral contract for electricity sales company at time t; and These represent the settlement prices for positive and negative deviations in electricity sales by the electricity sales company; α1 and α2 are 0 / 1 variables. When the deviation in electricity sales by the electricity sales company is positive, α1 = 1 and α2 = 0, otherwise α1 = 0 and α2 = 1. L represents the electricity price at time t in the market today; L represents the number of medium- and long-term contracts signed by the electricity retailers. This refers to the electricity volume stipulated in medium- and long-term contracts. Let be the proportion of the electrical energy allocated to contract l at time t.

2. The guidance method for flexible interaction of demand-side participation considering the impact of social information as described in claim 1, characterized in that: The impact of social information on user electricity load is quantified using the maximum information coefficient model. The maximum information coefficient model is... In the formula: MIC represents the correlation between social information and user electricity load; X represents social information; Y represents user electricity load; n x ,n y G represents the number of grid cells along the x and y axes; G is n. x ×n y The formed grid; I G (X,Y) represents the mutual information under grid G; B(n,α)=n α (0 < α < 1) is a function used to limit the maximum number of cells; log2min(n x ,n y ) is a standardization item that ensures that MIC is in the range of 0 to 1.

3. The guidance method for flexible interaction of demand-side participation considering the impact of social information as described in claim 1, characterized in that: Calculate user electricity consumption using the Attention-LSTM load prediction model: x t =[Te j (t),h j (t),p j (t),k j (t),q j (t-1),q j (t-2),q j-1 (t),q j-1 (t-1)] In the formula: f represents the user's electricity consumption at time t; t Here is the electricity load prediction function; x t Let be the input feature vector at time t; j represents the number of days. Te j (t) represents the predicted temperature at the predicted time, h j (t) represents the predicted humidity and p at the predicted time. j (t) represents the electricity price at the predicted time, k j (t) represents the coupon coefficient at the predicted time point, q j (t-1) represents the load value at the time before the prediction point, q j (t-2) represents the load value two moments before the prediction point, q j-1 (t) represents the load value at the same time the day before the prediction point, q j-1 (t-1) represents the load value at the time one day before the prediction point.

4. The guidance method for flexible interaction of demand-side participation considering the impact of social information as described in claim 1, characterized in that: The multi-tiered market power purchase model for electricity retailers includes a medium- and long-term market power purchase model and a day-ahead market power purchase model, respectively. In the formula: B Y For the cost of purchasing electricity in the medium to long term market; B D Y represents the cost of electricity purchase in the day-ahead market; D represents the cost in the medium- and long-term market; L represents the number of medium- and long-term contracts signed by electricity retailers. This refers to the electricity volume stipulated in medium- and long-term contracts. The price of a medium- to long-term contract; The amount of electricity purchased at time t in the market before the current date; The electricity price at time t in the market is as follows.

5. The guidance method for flexible interaction of demand-side participation considering the impact of social information as described in claim 1, characterized in that: The process of solving the optimal decision-making model for electricity purchase and sale by e-commerce platforms includes the following steps: Step 1: Input the parameters required for the e-commerce e-commerce purchase and sales optimization decision model; Step 2: Set the number of particles and the maximum number of iterations, where each particle represents a coupon coefficient; Step 3: Randomly generate an initial particle swarm, calculate the fitness value of all individuals in the initial particle swarm, and obtain the individual extreme values ​​and the global extreme value. Step 4: Each particle follows two extreme values ​​to change its velocity and position, and then compares them with the individual extreme value and the global extreme value. If it is better than the individual extreme value or the global extreme value, the individual extreme value or the global extreme value is updated, and the position and velocity of each particle are updated with the updated individual extreme value and the global extreme value to form a new particle. Step 5: Call the user electricity consumption behavior analysis program and use Attention-LSTM to solve the electricity consumption of each user at each time based on the new particles; Step 6: Call the electricity sales optimization subroutine, and use the CPLEX solver to solve for the electricity sales revenue of the electricity sales platform based on the electricity sales platform's electricity purchase and sale optimization decision model, with the objective function of maximizing benefits. Step 7: Increment the current iteration count by one and compare it with the set maximum iteration count. If the maximum iteration count has not been reached, return to step 4; otherwise, end the iteration process and determine the optimal demand-side adjustable load using the particle updated in the last iteration count.

6. The guidance method for flexible interaction of demand-side participation considering the impact of social information as described in claim 1, characterized in that: The optimal demand-side adjustable load is the electricity consumption of a user at a corresponding time of day, determined based on the optimal coupon coefficient, and users are guided to participate in load regulation based on the electricity consumption.

7. A system for implementing a guidance method for flexible interaction of demand-side participation that considers the impact of social information as described in any one of claims 1-6, characterized in that: include The data acquisition module is used to collect social information data; The user electricity consumption prediction module determines the degree of impact of each piece of social information on the user's electricity load, and determines the user's electricity consumption based on the social information corresponding to the degree of impact. The model building module is used to build a multi-level market electricity purchase model for electricity retailers, and to build an optimized decision-making model for electricity purchase and sales by electricity retailers based on social information, user electricity consumption, and the multi-level market electricity purchase model of electricity retailers. The flexible interaction module is used to solve the electricity purchase and sale optimization decision model of the e-commerce platform using the particle swarm optimization algorithm and the CPLEX solver to obtain the optimal demand-side adjustable load, and guide users to participate in flexible interaction based on the optimal demand-side adjustable load.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 6.

Images