User-side time-of-use electricity price optimization method and device for new energy consumption in rural area

Abstract

The invention discloses a rural area new energy consumption-oriented user side time-of-use electricity price optimization method, which comprises the steps of determining an original daily net load curve according to user side load data and new energy output data of a rural area; clustering the original daily net load curve by using an improved K-means + + algorithm to obtain a typical daily net load curve; carrying out peak-valley period refined division based on the typical daily net load curve to obtain a peak-valley period division result; taking the minimum peak-valley difference rate of the power system in the rural area as a target, constructing a target function with the minimum peak-valley difference rate, and constructing a time-of-use electricity price optimization model in combination with the peak-valley time period division result; and solving the time-of-use electricity price optimization model by using an RODDPSO algorithm to obtain an optimal time-of-use electricity price scheme. Based on this, the method can effectively stimulate the user-side adjustable load to participate in demand response, guides the time-space transfer of the load, and smooths the rural power grid net load curve while improving the new energy consumption.

Description

Technical Field

[0001] This invention relates to the field of power technology, and in particular to a user-side time-of-use pricing optimization method and device for new energy consumption in rural areas. Background Technology

[0002] The rural power grid infrastructure is relatively weak. The inherent intermittency and volatility of renewable energy generation, combined with the characteristics of rural electricity load, creates new problems of source-load mismatch in time and space. This leads to severe curtailment of solar power during midday due to excess distributed photovoltaic power generation that cannot be absorbed, while during morning and evening peak hours, the power supply must rely on the upstream grid, exacerbating the operational pressure on the rural power grid.

[0003] Currently, my country's time-of-use (TOU) pricing system uses a fixed peak-valley time period division throughout the year, failing to consider the dynamic characteristics of rural renewable energy output and rural load. This results in a mismatch between the time period division and the actual net load curve in rural areas. Furthermore, the existing TOU pricing mechanisms are mostly provincially unified, failing to differentiate pricing based on the actual conditions of rural power grids and the electricity consumption characteristics of different types of rural load users. Due to the unreasonable time period division and price differences, the TOU pricing mechanism cannot effectively promote the consumption of renewable energy in rural areas, nor can it effectively incentivize rural users to participate in demand response.

[0004] Therefore, a user-side time-of-use pricing optimization method for renewable energy consumption in rural areas is needed to address the problems existing in the above-mentioned technical solutions. Summary of the Invention

[0005] To address this, the present invention provides a user-side time-of-use pricing optimization method and device for new energy consumption in rural areas, in order to solve or at least alleviate the problems mentioned above.

[0006] According to one aspect of the present invention, a user-side time-of-use pricing optimization method for renewable energy consumption in rural areas is provided, executed in a computing device, comprising: acquiring multi-dimensional power data of rural areas, the multi-dimensional power data including renewable energy output data, user-side load data, and historical electricity price data; determining an original daily net load curve based on the user-side load data and the renewable energy output data; clustering the original daily net load curve using an improved K-means++ algorithm to obtain a typical daily net load curve, the typical daily net load curve including the net load of each time period of the typical day; refining the peak-valley time period based on the typical daily net load curve to obtain peak-valley time period division results, the peak-valley time period division results including peak time period, high-peak time period, normal time period, low-valley time period, and deep-valley time period; and optimizing the rural power system. With the goal of minimizing the peak-valley difference rate, a peak-valley difference rate objective function is constructed. Based on this objective function and the peak-valley time period division results, a time-of-use pricing optimization model is built. The peak-valley difference rate represents the ratio of the difference between the maximum and minimum loads in each time period after time-of-use pricing optimization to the maximum load. The loads in each time period after time-of-use pricing optimization are determined based on the original loads and original electricity prices of each time period before optimization, the electricity prices of each time period after optimization, and the elasticity coefficient of electricity demand in each time period to the change in electricity prices. The RODDPSO algorithm is used to solve the time-of-use pricing optimization model to obtain the optimal time-of-use pricing scheme. The optimal time-of-use pricing scheme indicates the optimal peak-hour electricity price, mid-peak-hour electricity price, normal-hour electricity price, low-valley-hour electricity price, and deep-valley-hour electricity price.

[0007] Optionally, in the user-side time-of-use pricing optimization method for new energy consumption in rural areas according to the present invention, the new energy output data includes the new energy power generation in each time period before the time-of-use pricing optimization, the user-side load data includes the original load in each time period before the time-of-use pricing optimization, and the historical electricity price data includes the original electricity price in each time period before the time-of-use pricing optimization.

[0008] Optionally, in the user-side time-of-use electricity price optimization method for new energy consumption in rural areas according to the present invention, a time-of-use electricity price optimization model is constructed based on the objective function of minimizing the peak-valley difference rate and in combination with the peak-valley time period division results, including: constructing an elasticity matrix, wherein the elasticity matrix includes the elasticity coefficient of electricity price changes in each time period on the electricity demand in each time period; and constructing a time-of-use electricity price optimization model based on the objective function of minimizing the peak-valley difference rate and the elasticity matrix, in combination with the peak-valley time period division results.

[0009] Optionally, in the user-side time-of-use electricity price optimization method for new energy consumption in rural areas according to the present invention, a time-of-use electricity price optimization model is constructed based on the objective function of minimizing the peak-valley difference rate and the elasticity matrix, combined with the peak-valley time period division results. This includes: constructing a time-of-use electricity price optimization model based on the objective function of minimizing the peak-valley difference rate, the elasticity matrix, and multiple constraints, combined with the peak-valley time period division results. The multiple constraints include: a rural user electricity cost constraint, indicating that the total electricity cost paid by rural users after time-of-use electricity price optimization is not higher than the total electricity cost paid by rural users before time-of-use electricity price optimization; and a peak-valley time period electricity price constraint, indicating that the electricity price during peak hours is greater than that during off-peak hours. Electricity price is greater than the normal period price, which is greater than the off-peak period price, which is greater than the deep off-peak period price; Peak-valley load constraints before and after optimization indicate that the peak load before optimization is greater than or equal to the peak load after optimization, and the off-peak load before optimization is less than or equal to the off-peak load after optimization; Rural user regulation capacity constraints indicate that the change in adjustable load of rural users is less than or equal to the proportion of adjustable load; Electricity balance constraints indicate that the total daily load after optimization is equal to the total daily load before optimization; Renewable energy consumption constraints indicate that the proportion of renewable energy generation in rural areas to the total electricity consumption of the power system is greater than or equal to the minimum renewable energy consumption ratio.

[0010] Optionally, in the user-side time-of-use pricing optimization method for new energy consumption in rural areas according to the present invention, the net load of each time period in the original daily net load curve is taken as a sample point; the original daily net load curve is clustered using the improved K-means++ algorithm to obtain a typical daily net load curve, including: traversing each candidate K value within a predetermined search range, and performing the following steps for each candidate K value: randomly selecting a sample point from the original daily net load curve as the first cluster center; for each sample point in the original daily net load curve, calculating the shortest distance between the sample point and the first cluster center, and determining the probability that the sample point is selected as the next cluster center based on the shortest distance; selecting the next cluster center based on the probability that each sample point is selected as the next cluster center, until K cluster centers are obtained. Using the K-means algorithm, based on the original daily net load curve and the K cluster centers, K clusters are formed. For each sample point in the original daily net load curve, the silhouette coefficient of the sample point is calculated based on the average distance between the sample point and other sample points in its cluster, and the average distance between the sample point and all points in its nearest neighbor cluster. Based on the silhouette coefficient of each sample point, the average silhouette coefficient of all sample points in the original daily net load curve is calculated as the average silhouette coefficient corresponding to the candidate K value. The maximum average silhouette coefficient among the average silhouette coefficients corresponding to each candidate K value is determined, and the candidate K value corresponding to the maximum average silhouette coefficient is determined as the optimal K value. The optimal K value and the cluster centers under the optimal K value are used as the final clustering result, and a typical daily net load curve is obtained based on the final clustering result.

[0011] Optionally, in the user-side time-of-use pricing optimization method for new energy consumption in rural areas according to the present invention, the K-means algorithm is used to form K-value clusters based on the original daily net load curve and the K-value cluster centers. The steps include: for each sample point in the original daily net load curve, assigning the sample point to the cluster center closest to the sample point among the K-value cluster centers to form K-value clusters; redetermining the new cluster center of each cluster, including: calculating the mean of all sample points in each cluster as the new cluster center of each cluster; repeating the above steps until the cluster center of each cluster no longer changes or the maximum number of iterations is reached, to obtain the final K-value cluster centers and K-value clusters.

[0012] Optionally, in the user-side time-of-use electricity price optimization method for new energy consumption in rural areas according to the present invention, the peak-valley time period is refined based on the typical daily net load curve, including: setting a minimum time period division constraint factor based on the typical daily net load curve and using the minimum DBI value as the objective function to construct a peak-valley time period division model; and refining the peak-valley time period division based on the peak-valley time period division model.

[0013] Optionally, in the user-side time-of-use pricing optimization method for new energy consumption in rural areas according to the present invention, the RODDPSO algorithm is used to solve the time-of-use pricing optimization model to obtain the optimal time-of-use pricing scheme, including the following steps: initializing algorithm parameters, the algorithm parameters including particle swarm size, maximum particle velocity, minimum particle velocity, and maximum number of iterations, wherein the particle position of each particle in the particle swarm is used to represent the time-of-use pricing scheme; based on the algorithm parameters and peak-valley time-of-use pricing constraints, randomly generating the initial particle position and initial particle velocity of each particle in the particle swarm; using the objective function of minimizing peak-valley difference, calculating the initial particle position and initial particle velocity of each particle in the particle swarm based on the particle position of each particle in the particle swarm. The fitness value of each particle is used to determine the particle swarm's position and velocity. Based on this fitness value, the swarm is divided into an exploration subgroup and a development subgroup, which are respectively adapted to use a global search strategy and a local development strategy to update particle positions and velocities. The above steps are iteratively executed, introducing random complementary solutions during the iteration process. Each iteration is performed a predetermined number of times, and particles are migrated between the exploration and development subgroups based on their fitness improvement rate to reorganize the swarm. When a termination condition is met, the optimal particle position in the swarm is taken as the optimal time-of-use pricing scheme. The termination condition includes reaching the maximum number of iterations or the particle positions no longer changing.

[0014] According to one aspect of the present invention, a user-side time-of-use electricity pricing optimization device is provided, deployed in a computing device, comprising: The acquisition module is suitable for acquiring multi-dimensional power data in rural areas, including new energy output data, user-side load data, and historical electricity price data. The determination module is adapted to determine the original daily net load curve based on the user-side load data and the new energy output data; A clustering module is adapted to cluster the original daily net load curve to obtain a typical daily net load curve, wherein the typical daily net load curve includes the net load for each time period of a typical day; The time period segmentation module is suitable for refining the peak and valley time periods based on the typical daily net load curve to obtain peak and valley time period segmentation results, which include peak time periods, high-peak time periods, normal time periods, low-peak time periods, and deep valley time periods. The module is suitable for constructing a peak-valley difference objective function with the goal of minimizing the peak-valley difference rate of the rural power system. Based on the peak-valley difference objective function and the peak-valley time period division results, a time-of-use pricing optimization model is constructed. The peak-valley difference rate represents the ratio of the difference between the maximum load and the minimum load in each time period after the time-of-use pricing optimization to the maximum load. The load in each time period after the time-of-use pricing optimization is suitable for determination based on the original load and original electricity price of each time period before the time-of-use pricing optimization, the electricity price of each time period after the time-of-use pricing optimization, and the elasticity coefficient of electricity demand in each time period to the change of electricity price in each time period. The solution module is adapted to solve the time-of-use electricity price optimization model to obtain the optimal time-of-use electricity price scheme, which is used to indicate the optimal peak period electricity price, high-peak period electricity price, normal period electricity price, low-peak period electricity price, and deep-valley period electricity price.

[0015] Optionally, in the user-side time-of-use pricing optimization device according to the present invention, the renewable energy output data includes renewable energy generation in each time period before time-of-use pricing optimization, the user-side load data includes the original load in each time period before time-of-use pricing optimization, and the historical electricity price data includes the original electricity price in each time period before time-of-use pricing optimization.

[0016] Optionally, in the user-side time-of-use electricity pricing optimization device according to the present invention, the construction module is adapted to construct a time-of-use electricity pricing optimization model based on the objective function of minimizing the peak-valley difference rate and in combination with the peak-valley time period division results in the following manner: constructing an elasticity matrix, the elasticity matrix including the elasticity coefficient of electricity price changes in each time period on the electricity demand in each time period; constructing a time-of-use electricity pricing optimization model based on the objective function of minimizing the peak-valley difference rate and the elasticity matrix, in combination with the peak-valley time period division results.

[0017] Optionally, in the user-side time-of-use electricity price optimization device according to the present invention, the construction module is further adapted to construct a time-of-use electricity price optimization model based on the objective function of minimizing the peak-valley difference rate and the elasticity matrix, combined with the peak-valley time period division results, in the following manner: Based on the objective function of minimizing the peak-valley difference rate, the elasticity matrix, and multiple constraints, combined with the peak-valley time period division results, a time-of-use electricity price optimization model is constructed, wherein the multiple constraints include: a rural user electricity cost constraint, indicating that the total electricity cost paid by rural users after time-of-use electricity price optimization is not higher than the total electricity cost paid by rural users before time-of-use electricity price optimization; and a peak-valley time period electricity price constraint, indicating that the electricity price during peak hours is greater than the peak hour price. The time-of-use (TOU) electricity price is greater than the normal time-of-use (TOU) price, which is greater than the low-peak time-of-use (SHN) price, which is greater than the deep-peak time-of-use (STN) price. The optimization of peak-valley load constraints means that the peak load before TOU optimization is greater than or equal to the peak load after TOU optimization, and the valley load before TOU optimization is less than or equal to the valley load after TOU optimization. The rural user regulation capacity constraint means that the change in adjustable load of rural users is less than or equal to the proportion of adjustable load. The power balance constraint means that the total daily load after TOU optimization is equal to the total daily load before TOU optimization. The renewable energy consumption constraint means that the proportion of renewable energy generation in rural areas to the total electricity consumption of the power system is greater than or equal to the minimum renewable energy consumption ratio.

[0018] Optionally, in the user-side time-of-use pricing optimization device according to the present invention, the net load of each time period in the original daily net load curve is taken as a sample point; the clustering module includes: The traversal unit is suitable for traversing each candidate K value within a predetermined search range; The first selection unit is adapted to randomly select a sample point from the original daily net load curve as the first cluster center; The distance calculation unit is adapted to calculate the shortest distance between each sample point and the first cluster center for each sample point in the original daily net load curve, and determine the probability that the sample point is selected as the next cluster center based on the shortest distance. The second selection unit is adapted to select the next cluster center based on the probability that each sample point is selected as the next cluster center, until K cluster centers are obtained; The forming unit is suitable for using the K-means algorithm to form K clusters based on the original daily net load curve and the K cluster centers; The profile coefficient calculation unit is adapted to calculate the profile coefficient of each sample point in the original daily net load curve based on the average distance between the sample point and other sample points in its cluster, and the average distance between the sample point and all points in its nearest neighbor cluster. The average profile coefficient calculation unit is adapted to calculate the average profile coefficient of all sample points in the original daily net load curve based on the profile coefficient of each sample point, and use it as the average profile coefficient corresponding to the candidate K value. The determining unit is adapted to determine the maximum average profile coefficient among the average profile coefficients corresponding to each candidate K value, and to determine the candidate K value corresponding to the maximum average profile coefficient as the optimal K value. The clustering unit is adapted to take the optimal K value and each cluster center under the optimal K value as the final clustering result, and obtain a typical daily net load curve based on the final clustering result.

[0019] Optionally, in the user-side time-of-use pricing optimization device according to the present invention, the forming unit is adapted to use the K-means algorithm to form K-value clusters based on the original daily net load curve and the K-value cluster centers according to the following steps: for each sample point in the original daily net load curve, the sample point is assigned to the cluster center closest to the sample point among the K-value cluster centers to form K-value clusters; a new cluster center for each cluster is re-determined, including: calculating the mean of all sample points in each cluster as the new cluster center for each cluster; repeating the above steps until the cluster center of each cluster no longer changes or the maximum number of iterations is reached, to obtain the final K-value cluster centers and K-value clusters.

[0020] Optionally, in the user-side time-of-use pricing optimization device according to the present invention, the time period division module is adapted to refine the peak-valley time period division based on the typical daily net load curve in the following manner: based on the typical daily net load curve, a minimum time period division constraint factor is set and the minimum DBI value is used as the objective function to construct a peak-valley time period division model; and the peak-valley time period division is refined based on the peak-valley time period division model.

[0021] Optionally, in the user-side time-of-use electricity price optimization device according to the present invention, the solution module is adapted to solve the time-of-use electricity price optimization model according to the following steps to obtain the optimal time-of-use electricity price scheme: initializing algorithm parameters, the algorithm parameters including particle swarm size, maximum particle velocity, minimum particle velocity, and maximum number of iterations, wherein the particle position of each particle in the particle swarm is used to represent the time-of-use electricity price scheme; randomly generating the initial particle position and initial particle velocity of each particle in the particle swarm based on the algorithm parameters and peak-valley time-of-use electricity price constraints; and calculating the fitness value of each particle in the particle swarm based on the particle position of each particle in the particle swarm using the objective function of minimizing peak-valley difference. Based on the fitness value of each particle in the particle swarm, the particle swarm is divided into an exploration subgroup and a development subgroup. The exploration subgroup and the development subgroup are respectively adapted to use a global search strategy and a local development strategy to update particle positions and particle velocities. The above steps are executed iteratively, and random complementary solutions are introduced during the iteration process. Each iteration is executed a predetermined number of times, and particle migration is performed between the exploration subgroup and the development subgroup according to the fitness improvement rate of each particle in the particle swarm to reorganize the particle swarm. When the termination condition is met, the optimal particle position in the particle swarm is taken as the optimal time-of-use electricity pricing scheme, wherein the termination condition includes reaching the maximum number of iterations or the positions of each particle no longer changing.

[0022] According to one aspect of the present invention, a computing device is provided, comprising: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be executed by the at least one processor, the program instructions including instructions for executing the user-side time-of-use electricity pricing optimization method for new energy consumption in rural areas as described above.

[0023] According to one aspect of the present invention, a computer program product is provided, comprising computer program instructions, wherein the computer program instructions, when executed by a processor, implement the method as described above.

[0024] According to one aspect of the present invention, a readable storage medium storing program instructions is provided, which, when read and executed by a computing device, causes the computing device to perform the user-side time-of-use electricity pricing optimization method for new energy consumption in rural areas as described above.

[0025] According to the technical solution of the present invention, a user-side time-of-use electricity price optimization method and device for renewable energy consumption in rural areas are provided. The improved K-means++ algorithm is used to cluster the original daily net load curves of rural areas to obtain typical daily net load curves. Based on the typical daily net load curves, peak and valley periods are further subdivided into five categories: peak period, high-peak period, normal period, low-valley period, and deep-valley period. Then, a peak-valley difference objective function is constructed with the goal of minimizing the peak-valley difference rate of the rural power system. A time-of-use electricity price optimization model is constructed by combining the peak-valley period division results. Finally, the optimal time-of-use electricity price scheme is obtained by solving the time-of-use electricity price optimization model using the RODDPSO algorithm. Based on this, the present invention utilizes an improved K-means++ algorithm to automatically determine the optimal number of clusters, thereby improving the algorithm's convergence speed and clustering quality. It further refines the peak-valley periods based on typical daily net load curves, and on this basis, constructs a time-of-use pricing optimization model that considers the elasticity between rural users' electricity demand and electricity prices. The optimal time-of-use pricing scheme determined in this way can effectively incentivize adjustable loads on the user side to participate in demand response and guide the spatiotemporal transfer of loads, thus accurately matching the spatiotemporal characteristics of rural renewable energy output and load, solving the problem of source-load spatiotemporal mismatch, and smoothing the rural power grid net load curve while improving renewable energy absorption.

[0026] Furthermore, this invention constructs a peak-valley time period division model based on a typical daily net load curve, setting a minimum time period division constraint factor and using the minimum DBI value as the objective function. The peak-valley time period division is then refined based on this model. This allows for the dynamic determination of the time boundaries for five types of peak-valley time periods, resulting in a more reasonable division of peak-valley time periods.

[0027] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0028] To achieve the foregoing and related objectives, certain illustrative aspects are described herein in conjunction with the following description and accompanying drawings. These aspects indicate various ways in which the principles disclosed herein may be practiced, and all aspects and their equivalents are intended to fall within the scope of the claimed subject matter. The foregoing and other objectives, features, and advantages of the invention will become more apparent from the following detailed description, taken in conjunction with the accompanying drawings. Throughout the invention, the same reference numerals generally refer to the same parts or elements.

[0029] Figure 1 A schematic diagram of a computing device 100 provided according to an embodiment of the present invention is shown; Figure 2A flowchart illustrating a user-side time-of-use electricity pricing optimization method 200 for renewable energy consumption in rural areas, provided by an embodiment of the present invention, is shown. Figure 3 This diagram illustrates the monthly load curves of charging piles in a rural area of ​​a certain province, according to some embodiments of the present invention. Figure 4 and Figure 5 Schematic diagrams of the daily net load curve clustering results for January and February are shown respectively, according to some embodiments of the present invention; Figure 6 and Figure 7 The following are schematic diagrams showing the peak and valley period division results for January and February according to some embodiments of the present invention; Figure 8 A schematic diagram of a user-side time-of-use electricity pricing optimization device 800 provided according to an embodiment of the present invention is shown; Figure 9 A schematic diagram of a clustering module 830 according to some embodiments of the present invention is shown. Detailed Implementation

[0030] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0031] To address the issues of existing time-of-use (TOU) pricing mechanisms failing to match peak-valley periods with the actual net load curves in rural areas, thus hindering effective incentives for adjustable user-side loads to participate in demand response and promote renewable energy consumption in rural areas, this invention proposes a user-side TOU optimization method for renewable energy consumption in rural areas. Based on a typical daily net load curve, the peak-valley periods are further refined, and a TOU optimization model is constructed. This model enables precise, differentiated, and dynamic TOU pricing for rural areas, effectively incentivizing adjustable user-side loads to participate in demand response and guiding load transfer in time and space. This improves renewable energy consumption while smoothing the net load curve of the rural power grid.

[0032] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0033] Figure 1 A schematic diagram of a computing device 100 according to an embodiment of the present invention is shown. Figure 1As shown, in a basic configuration, computing device 100 includes at least one processing unit 102 and system memory 104. According to one aspect, depending on the configuration and type of the computing device, the processing unit 102 may be implemented as a processor. System memory 104 includes, but is not limited to, volatile memory (e.g., random access memory), non-volatile memory (e.g., read-only memory), flash memory, or any combination of such memories. According to one aspect, system memory 104 includes an operating system 105.

[0034] According to one aspect, operating system 105 is, for example, suitable for controlling the operation of computing device 100. Furthermore, examples are practiced in conjunction with graphics libraries, other operating systems, or any other applications, and are not limited to any particular application or system. Figure 1 The basic configuration is illustrated by the components within the dashed lines. According to one aspect, the computing device 100 has additional features or functions. For example, according to one aspect, the computing device 100 includes additional data storage devices (removable and / or non-removable), such as disks, optical discs, or magnetic tapes. This additional storage... Figure 1 The middle part is shown by removable storage device 109 and non-removable storage device 110.

[0035] As stated above, according to one aspect, program module 103 is stored in system memory 104. According to one aspect, program module 103 may include one or more applications. The present invention does not limit the type of application; for example, applications may include: email and contact applications, word processing applications, spreadsheet applications, database applications, slideshow applications, drawing or computer-aided applications, web browser applications, etc.

[0036] According to one aspect, program module 103 may include a plurality of program instructions adapted to execute the user-side time-of-use electricity price optimization method 200 for rural renewable energy consumption of the present invention, such that computing device 100 is configured to execute the user-side time-of-use electricity price optimization method 200 for rural renewable energy consumption of the present invention.

[0037] According to one aspect, examples can be practiced on circuits including discrete electronic components, packaged or integrated electronic chips containing logic gates, circuits utilizing microprocessors, or on a single chip containing electronic components or a microprocessor. For example, it can be practiced via wherein... Figure 1Each or many of the components shown can be implemented as an example by integrating a System-on-a-Chip (SOC) on a single integrated circuit. According to one aspect, such an SOC device may include one or more processing units, graphics units, communication units, system virtualization units, and various application functions, all integrated (or “burned in”) as a single integrated circuit onto a chip substrate. When operating via the SOC, the functions described herein can be operated via dedicated logic integrated on a single integrated circuit (chip) with other components of the computing device 100. Embodiments of the invention can also be implemented using other techniques capable of performing logical operations (e.g., AND, OR, and NOT), including but not limited to mechanical, optical, fluid, and quantum technologies. Additionally, embodiments of the invention can be implemented within a general-purpose computer or in any other circuit or system.

[0038] According to one aspect, computing device 100 may also have one or more input devices 112, such as a keyboard, mouse, pen, voice input device, touch input device, etc. It may also include output devices 114, such as a display, speaker, printer, etc. The foregoing devices are examples and other devices may also be used. Computing device 100 may include one or more communication connections 116 that allow communication with other computing devices 118. Examples of suitable communication connections 116 include, but are not limited to: RF transmitter, receiver and / or transceiver circuitry; Universal Serial Bus (USB), parallel and / or serial ports.

[0039] As used herein, the term computer-readable medium includes computer storage medium. Computer storage medium can include volatile and non-volatile, removable and non-removable media implemented using any method or technology for storing information (e.g., computer-readable instructions, data structures, or program module 103). System memory 104, removable storage device 109, and non-removable storage device 110 are examples of computer storage media (i.e., memory storage). Computer storage media can include random access memory (RAM), read-only memory (ROM), electrically erasable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical storage, magnetic tape, magnetic tape, disk storage or other magnetic storage devices, or any other article of manufacture that can be used to store information and is accessible by computing device 100. According to one aspect, any such computer storage medium can be part of computing device 100. Computer storage media does not include carrier waves or other transmitted data signals.

[0040] According to one aspect, the communication medium is implemented by computer-readable instructions, data structures, program modules 103, or other data in a modulated data signal (e.g., a carrier wave or other transmission mechanism), and includes any information transmission medium. According to one aspect, the term "modulated data signal" describes a signal having one or more sets of characteristics or altered in a manner that encodes information in the signal. By way of example and not limitation, the communication medium includes wired media such as wired networks or direct wired connections, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.

[0041] In an embodiment of the present invention, a computing device 100 is configured to execute the user-side time-of-use electricity pricing optimization method 200 for rural renewable energy consumption according to the present invention. The computing device 100 includes one or more processors and one or more readable storage media storing program instructions, which, when configured to be executed by the one or more processors, cause the computing device to execute the user-side time-of-use electricity pricing optimization method 200 for rural renewable energy consumption according to the embodiments of the present invention.

[0042] Figure 2 A flowchart illustrating a user-side time-of-use pricing optimization method 200 for renewable energy consumption in rural areas, according to an embodiment of the present invention, is shown. The user-side time-of-use pricing optimization method 200 for renewable energy consumption in rural areas can be executed in a computing device (e.g., the aforementioned computing device 100).

[0043] In an embodiment of the present invention, the computing device 100 used to execute the user-side time-of-use electricity price optimization method 200 for new energy consumption in rural areas can be a terminal or a server.

[0044] like Figure 2 As shown, the user-side time-of-use electricity pricing optimization method 200 for new energy consumption in rural areas includes the following steps 210-260.

[0045] Step 210: The computing device 100 can acquire multi-dimensional power data of rural areas before time-of-use electricity pricing optimization. The multi-dimensional power data may include renewable energy output data (distributed renewable energy output data), user-side load data, and historical electricity price data.

[0046] It should be noted that the renewable energy output data includes renewable energy generation (total wind and solar power generation) for each period before the time-of-use pricing optimization. User-side load data includes the original load for each period before the time-of-use pricing optimization. Historical electricity price data includes the original electricity price for each period before the time-of-use pricing optimization.

[0047] Step 220: The calculation device 100 can determine the original daily net load curve of rural areas based on user-side load data (original load of each period before time-of-use pricing optimization) and renewable energy output data (renewable energy power generation of each period before time-of-use pricing optimization).

[0048] It should be noted that the original daily net load curve includes the net load for each time period before the optimization of time-of-use pricing in rural areas. This original daily net load curve can accurately reflect the supply and demand balance of the power system in rural areas.

[0049] In some embodiments, before performing step 220, abnormal data in the user-side load data and new energy output data can be removed in advance, and then the original daily net load curve can be calculated.

[0050] In some embodiments, the original daily net load curve is shown in the following formula: (1) In the formula, express Net load during the period; express User-side load during time of day (i.e., before time-of-use pricing optimization) (Original load for the time period) express Renewable energy power generation during the period.

[0051] Step 230: The computing device 100 can cluster the original daily net load curves to obtain typical daily net load curves.

[0052] In some embodiments, in step 230, the computing device 100 may use an improved K-means++ algorithm to cluster the original daily net load curve to obtain a typical daily net load curve.

[0053] It should be noted that the typical daily net load curve includes the net load for each time period of a typical day. Using the improved K-means++ algorithm, multi-dimensional cluster analysis can be performed on the original daily net load curve at the quarterly and monthly levels. Ultimately, this can effectively output typical daily net load curves for different seasons and months in rural areas.

[0054] In some embodiments, the time period can be an hour; that is, the typical daily net load curve can include the net load for each hour of a typical day.

[0055] In some embodiments, the original daily net load curves may be preprocessed before clustering.

[0056] Step 240: The computing device 100 can refine the peak-valley time period division based on the typical daily net load curve to obtain the peak-valley time period division results for the rural power system. Specifically, an improved moving boundary technique can be used to refine the peak-valley time period division based on the typical daily net load curve.

[0057] It should be understood that the detailed division of peak and valley periods based on the typical daily net load curve takes into account the rural load characteristics and the output of new energy sources, so that the division of peak and valley periods matches the actual net load curve in rural areas.

[0058] In this embodiment of the invention, the peak-valley time period division results include peak periods, high-peak periods, normal periods, low-valley periods, and deep-valley periods. It should be noted that this embodiment of the invention, combined with rural load characteristics, further refines the peak-valley time period division into five categories: peak, high-peak, normal, low, and deep-valley, improving the accuracy of peak-valley time period division and making the time period division more reasonable for rural areas. Specifically, peak periods correspond to the rural load peak coinciding with low renewable energy output (such as the evening peak hours for residential electricity consumption and agricultural processing); deep-valley periods correspond to the peak renewable energy generation coinciding with low load (such as the midday peak photovoltaic output and the time when residents are away from home); high-peak periods are suitable for the concentrated electricity consumption during busy farming seasons; and low-valley periods are suitable for the secondary peak renewable energy generation periods.

[0059] Step 250: The computing device 100 constructs a peak-valley difference objective function with the goal of minimizing the peak-valley difference rate in the rural power system. Then, based on this objective function and the peak-valley time period division results, a time-of-use pricing optimization model can be constructed. The peak-valley difference rate represents the ratio of the difference between the maximum and minimum load in each time period after time-of-use pricing optimization to the maximum load.

[0060] In this embodiment of the invention, the load for each time period after time-of-use pricing optimization can be determined based on the original load and original electricity price for each time period before optimization, the electricity price for each time period after optimization, and the elasticity coefficients (including self-elasticity coefficients and cross-elasticity coefficients) of the electricity demand for each time period affected by changes in electricity prices. The elasticity coefficients represent the impact of electricity price changes on user electricity demand. Based on this, a time-of-use pricing optimization model that considers the elasticity between user electricity demand and electricity prices can be constructed.

[0061] It should be noted that the elasticity coefficient includes the self-elasticity coefficient and the cross-elasticity coefficient. The self-elasticity coefficient represents the impact of electricity price changes in different time periods on user electricity demand during those periods. The cross-elasticity coefficient represents the impact of electricity price changes in different time periods on user electricity demand during other time periods.

[0062] Step 260: The computing device 100 can solve the time-of-use electricity pricing optimization model to obtain the optimal time-of-use electricity pricing scheme for rural areas. The optimal time-of-use electricity pricing scheme is used to indicate the optimal peak-hour electricity price, normal-hour electricity price, off-peak-hour electricity price, and deep-off-peak-hour electricity price for rural areas.

[0063] In some embodiments, in step 260, the computing device 100 may use the RODDPSO algorithm (Stochastic Distributed Delayed Particle Swarm Optimization) to solve the time-of-use electricity pricing optimization model to obtain the optimal time-of-use electricity pricing scheme for rural areas.

[0064] According to some embodiments of the present invention, in step 230, the specific process of clustering the original daily net load curve using the improved K-means++ algorithm to obtain a typical daily net load curve includes the following steps C1 to C9. Wherein, the number of clusters is represented by K.

[0065] C1. The search range of the K value can be pre-set to a predetermined search range. , The K value can be expressed as Iterate through each candidate K value within the predetermined search range. ( For each candidate K value, perform the following steps C2~C7: C2. Randomly select a sample point from the original daily net load curve as the first cluster center.

[0066] C3. For each sample point in the original daily net load curve, calculate the shortest distance between the sample point and the first cluster center, and then determine the probability that the sample point will be selected as the next cluster center based on the shortest distance.

[0067] Specifically, the sample points can be calculated using the following formula. Shortest distance to the first cluster center: (2) in, .

[0068] The probability of a sample point being selected as the next cluster center can be calculated using the following formula: (3) C4. Based on the probability (probability distribution) of each sample point being selected as the next cluster center, select the next cluster center and repeat steps C3-C4 until the K value is obtained. There are 100 cluster centers. The cluster centers obtained based on the probability distribution can maximize the distance between the cluster centers, which can improve the convergence speed and clustering quality compared to the traditional K-means algorithm.

[0069] C5. Using the K-means algorithm (i.e., the standard K-means algorithm), based on the original daily net load curve and K-value cluster centers, form K-value clusters.

[0070] Specifically, step C5 may include the following steps C51-C52: C51. For each sample point in the original daily net load curve, assign the sample point to the nearest cluster center among the K-value cluster centers to form K-value clusters. C52. Redetermine the new cluster centers for each cluster. Specifically, calculate the mean of all sample points within each cluster as the new cluster center for each cluster. Repeat steps C51-C52 until the cluster centers of each cluster no longer change (convergence) or the maximum number of iterations is reached, obtaining the final K-value cluster centers and K-value clusters. Based on the final K-value cluster centers and K-value clusters, perform the following steps.

[0071] C6. For each sample point in the original daily net load curve, the profile coefficient of that sample point can be calculated based on the average distance between that sample point and other sample points within its cluster, and the average distance between that sample point and all points within its nearest neighbor cluster. The specific formula for calculating the profile coefficient of a sample point is as follows: (4) In the formula, Represents the i-th sample point The average distance to other sample points within the same cluster; Represents the i-th sample point The average distance to all points in the nearest neighbor cluster.

[0072] C7. Based on the profile coefficient of each sample point, calculate the average profile coefficient of all sample points in the original daily net load curve, and use it as the above candidate K value. The corresponding average profile coefficient. Candidate K values ​​( The corresponding average profile coefficient is shown in the following formula: (5) In the formula, This represents the total number of sample points.

[0073] After traversing all candidate K values ​​within the predetermined search range and executing steps C2 to C7 above, step C8 can be executed to determine the optimal K value.

[0074] C8. Determine the maximum average profile coefficient among the average profile coefficients corresponding to each candidate K value within the predetermined search range, and determine the candidate K value corresponding to the maximum average profile coefficient as the optimal K value (i.e., the optimal number of clusters).

[0075] C9. The optimal K value ( The cluster centers at the optimal K value are used as the final clustering result, and a typical daily net load curve can be obtained from the final clustering result.

[0076] As can be seen, this invention, based on the improved K-means++ algorithm, can automatically determine the optimal number of clusters, ensuring the global optimality of the clustering results. Furthermore, compared with the traditional K-means algorithm, it can improve the convergence speed and clustering quality.

[0077] In some embodiments, in step 240, a peak-valley time period segmentation model can be constructed based on a typical daily net load curve, setting a minimum time period segmentation constraint factor and using the minimum DBI (Davidson-Bolding Index) value as the objective function. Furthermore, the peak-valley time period segmentation can be further refined based on this model. The specific process is as follows: First, you can input a typical daily net load curve. ,in Indicates the first Hourly net load, The number of time periods (hours) ).

[0078] Subsequently, the typical daily net load curves can be sorted in ascending order to obtain the ascending-order typical daily net load curves, i.e. ,in This represents the minimum load in the typical daily net load curve in ascending order. This represents the maximum load in the typical daily net load curve in ascending order.

[0079] Subsequently, decision variables and storage variables can be initialized. Specifically, let... are decision variables. Among them, The peak load is during the valley period; This represents the maximum load during normal periods. and These represent the first and second typical daily net load curves in ascending order. The load and the first A load, let , , , ; This represents the minimum time period segmentation constraint factor in the refined segmentation of peak and valley periods.

[0080] Furthermore, a minimum time period division constraint factor can be set, and the objective function can be the minimum DBI value of the decision variable to construct a peak-valley time period division model as follows: (6) (7) (8) (9) (10) In the formula, Indicates the first The number of net load points in each time period; Indicates the types of peak and valley periods that need to be divided; This represents the average value of all net load points within the same time period; Indicates the first Net load point for each time period and The standard error; express and The distance between them.

[0081] Then, the decision variables can be iteratively updated based on the following conditions. and Value: If Then let Otherwise , Until When the iteration to update decision variables is terminated, the process is stopped. and The value of .

[0082] Finally, output the minimum DBI value and its corresponding value. , .

[0083] Therefore, based on the peak-valley time period division model, the peak-valley time period can be further divided into five types of time periods (peak period, peak period, normal period, low period, and deep valley period).

[0084] In some embodiments, in step 250, the objective function for minimizing the peak-valley difference rate (the ratio of the difference between the maximum load and the minimum load in each time period after time-of-use pricing optimization to the maximum load) of the rural power system is constructed as shown in the following formula.

[0085] (11) The specific calculation formulas for load in each time period after time-of-use pricing optimization are as follows: (12) (13) In the formula, This indicates the optimization of time-of-use electricity pricing. Time-of-use load (user-side load). Indicates the periodic pricing before optimization The original load for the time period; and These represent the electricity prices before and after time-of-use pricing optimization. Time period and The original electricity price for that time period. The self-elasticity coefficient represents... The impact of changes in electricity prices during a given time period on users' electricity demand during that period; The cross-elasticity coefficient represents... Time-of-use electricity price changes The impact of user electricity demand during different time periods. Indicates the optimization of time-of-use electricity pricing Time-of-use electricity pricing This indicates the optimization of time-of-use electricity pricing. Electricity pricing for specific time periods.

[0086] In some embodiments, to more comprehensively and systematically describe the elasticity relationship between user electricity demand and electricity prices, an elasticity matrix (including the elasticity coefficients of electricity price changes on electricity demand in each time period) can be constructed. This matrix can then be used to represent the elasticity coefficients (including self-elasticity coefficients and cross-elasticity coefficients) of electricity price changes on electricity demand in each time period. Assuming the rural power system is divided into... Each time period, elastic matrix It is A matrix, in which the elements of the elasticity matrix ( )express Time-of-use electricity price changes The elasticity coefficient of electricity demand during different time periods. Elasticity matrix. Specifically as follows: (14) In the aforementioned elasticity matrix, the elements on the main diagonal... The self-elasticity coefficient is an element located on a non-main diagonal. This is the cross-elasticity coefficient.

[0087] Furthermore, a time-of-use electricity pricing optimization model can be constructed based on the objective function of minimizing the peak-valley difference rate and the elasticity matrix, combined with the peak-valley time period division results.

[0088] In addition, in some embodiments, multiple constraints need to be considered when constructing the time-of-use electricity pricing optimization model. That is, the time-of-use electricity pricing optimization model can be constructed based on the objective function of minimizing the peak-valley difference rate, the elasticity matrix, and multiple constraints, combined with the peak-valley time period division results.

[0089] Specifically, the constraints include rural user electricity cost constraints, peak-valley electricity price constraints, peak-valley load constraints before and after optimization, rural user regulation capacity constraints, power balance constraints, and renewable energy consumption constraints.

[0090] The constraint on electricity costs for rural users indicates that the total electricity cost paid by rural users after the optimization of time-of-use pricing should not be higher than the total electricity cost paid by rural users before the optimization, as shown in the following formula: (15) The peak-valley electricity pricing constraint is as follows: Peak-hour electricity price > High-peak-hour electricity price > Normal-hour electricity price > Low-valley-hour electricity price > Deepest-valley-hour electricity price. The specific formula is as follows: (16) In the formula, These represent the electricity price during peak hours, normal hours, off-peak hours, and deep off-peak hours, respectively.

[0091] The load constraints before and after optimization are expressed as follows: the peak load before time-of-use pricing optimization is greater than or equal to the peak load after time-of-use pricing optimization; and the valley load before time-of-use pricing optimization is less than or equal to the valley load after time-of-use pricing optimization. Specifically, the following formula is used: (17) In the formula, and These represent peak loads before and after time-of-use pricing optimization, respectively. and These represent the off-peak load before and after time-of-use pricing optimization, respectively.

[0092] The rural user adjustment capacity constraint states that the change in adjustable load for rural users is less than or equal to the proportion of adjustable load, as shown in the following formula. Based on this, the adjustable load proportions for different types of rural users are considered to ensure that the adjustable load for users participating in demand response is within their actual response capacity.

[0093] (18) In the formula, Indicates the amount of adjustable load change; This indicates the percentage of adjustable load.

[0094] The power balance constraint states that the total daily load after time-of-use pricing optimization (the sum of loads during all time periods within a day) equals the total daily load before time-of-use pricing optimization. Specifically, it is shown in the following formula: (19) The renewable energy consumption constraint means that the proportion of renewable energy power generation in rural areas to the total electricity consumption of the power system must be greater than or equal to the minimum renewable energy consumption ratio. The specific formula is as follows: (20) In the formula, This represents the total electricity consumption of the power system. This indicates the amount of electricity generated from new energy sources. This indicates the minimum proportion of renewable energy that can be absorbed.

[0095] It should be noted that in the scenario of time-of-use electricity pricing optimization in rural areas, optimization algorithms play an important role in solving energy data problems that are high-dimensional, nonlinear, and highly dynamic. RODDPSO (Random Oppositional Dimensionally Dynamic Particle Swarm Optimization), as an improved particle swarm optimization algorithm, significantly enhances the global search capability and local optimization efficiency of the traditional PSO algorithm by introducing a dynamic decision-making and adjustment mechanism based on randomized opposing dimensions. It is suitable for complex optimization scenarios in rural renewable energy systems.

[0096] In some embodiments, the specific process of solving the time-of-use electricity price optimization model using the RODDPSO algorithm (Stochastic Distributed Delayed Particle Swarm Optimization Algorithm) in step 260 includes the following steps F1 to F7.

[0097] F1. Initialize algorithm parameters. Algorithm parameters include particle swarm size, maximum particle velocity, minimum particle velocity, and maximum number of iterations. It should be noted that the particle positions in the particle swarm are used to represent the time-of-use pricing scheme.

[0098] In one specific embodiment, based on the characteristics of the time-of-use electricity pricing optimization problem in rural areas, the basic parameters of the particle swarm optimization are set as follows: particle swarm size sizepop=20, each particle is a 5-dimensional vector representing the electricity price during peak hours, normal hours, low hours, and deep low hours, respectively. The maximum particle velocity V_max=0.02, the minimum particle velocity V_min=-0.02, and the maximum number of iterations maxgen=100 are set. To balance the global exploration and local exploitation capabilities of the algorithm, the dynamic variation range of the inertia weight is set to w=[0.4,0.9], and the acceleration coefficients C1=2.0 and C2=2.0 are initialized.

[0099] F2. Based on algorithm parameters and peak / valley electricity price constraints, the initial particle position and initial particle velocity of each particle in the particle swarm can be randomly generated. Specifically, under the premise of satisfying the peak / valley electricity price constraints, the initial particle position is randomly generated; at the same time, the particle velocity is randomly initialized within the velocity range (between the minimum particle velocity and the maximum particle velocity) to generate the initial particle velocity.

[0100] F3. Using the objective function of minimizing peak-valley difference, calculate the fitness value (e.g., initial fitness value) of each particle in the particle swarm based on the particle position (e.g., initial particle position) of each particle in the particle swarm. At the same time, the constraints of electricity cost for rural users, peak-valley electricity price, and rural user regulation capacity should be considered.

[0101] F4. Based on the fitness value of each particle in the particle swarm, the particle swarm is dynamically divided into an exploration subgroup and an exploit subgroup. The exploration subgroup and the exploit subgroup are respectively adapted to use a global search strategy and a local exploit strategy to update the particle position and particle velocity.

[0102] F5. Iteratively execute steps F3~F4 above to iteratively optimize the time-of-use electricity pricing scheme corresponding to each particle position. Specifically: using the objective function of minimizing peak-valley difference rate, calculate the new fitness value of each particle in the particle swarm based on the updated particle position of each particle in the particle swarm; according to the new fitness value of each particle in the particle swarm, re-divide the particle swarm into an exploration subgroup and a development subgroup.

[0103] Furthermore, random opposing solutions can be introduced during the iteration process to enhance particle swarm diversity and prevent the algorithm from converging prematurely.

[0104] In some embodiments, a combination of feasibility rules and dynamic penalty functions can be used to ensure that the iterative process always takes place within the solution space that satisfies all constraints. F6. Execute a predetermined number of times per iteration (e.g., 10 times per iteration), and particle migration can be performed between the explore and develop subgroups based on the fitness improvement rate of each particle in the particle swarm to reorganize the particle swarm.

[0105] F7. When the termination condition is met, the optimal particle position in the particle swarm is taken as the optimal time-of-use electricity pricing scheme. The termination condition includes reaching the maximum number of iterations or the particle positions no longer changing. In other words, when the maximum number of iterations is reached or the particle positions no longer change, the algorithm terminates, and the optimal particle position in the particle swarm at this time can be output as the optimal time-of-use electricity pricing scheme.

[0106] In addition, in some embodiments, to verify the effectiveness of the model constructed by the present invention, a case study was conducted on an agricultural demonstration county in a province in central my country, and the monthly charging price of its electric vehicle charging piles was optimized. Figure 3 A schematic diagram of the monthly load curves of charging piles in a rural area of ​​a certain province is shown, according to some embodiments of the present invention. Additionally, relevant monthly indicators of electric vehicle charging piles in the rural area of ​​the province are shown in Table 1.

[0107] Table 1 Typical load characteristic indicators of charging piles (load unit: 10,000 kilowatts) To formulate a reasonable monthly time-of-use electricity pricing strategy, the improved K-means++ algorithm is first used to cluster the daily net load curves for each month. Due to the large number of months, January and February are used as examples in some implementations. Figure 4 and Figure 5 The diagrams show the clustering results of the daily net load curves for January and February according to some embodiments of the present invention. In the diagrams, the gray curve represents the daily net load curve (original daily net load curve) for that month, and the red curve represents the typical daily net load curve.

[0108] Building upon this foundation, the moving boundary technique is further employed to refine the peak and valley periods. Combining the monthly net load curve clustering results (typical daily net load curve), in addition to setting peak, flat, and valley periods, periods with significantly higher net load values ​​are selected as peak periods, and periods with significantly lower net load values ​​are selected as valley periods. Figure 6 and Figure 7 The diagrams showing the peak and valley period division results for January and February according to some embodiments of the present invention are illustrated. Table 2 shows the peak and valley period division for each month.

[0109] Table 2. Monthly Peak and Valley Period Division It should be noted that, due to the large scale of renewable energy power generation, the net load is generally low in the afternoon, so it is set as the low-load period or valley period; while the photovoltaic output is weak at night and the net load is high, so it is set as the peak period or peak period.

[0110] Based on the above peak and valley time period division, the monthly time-of-use electricity price for the electric vehicle industry was further optimized, and the resulting peak and valley coefficients are shown in Table 3.

[0111] Table 3 Peak-Valley Coefficient of Monthly Time-of-Use Electricity Price for Electric Vehicles In some embodiments, to evaluate the superiority of the time-of-use pricing optimization model proposed in this invention, the total load increase rate during off-peak hours is introduced. and peak-hour total load reduction rate The two indicators are calculated using the following formulas: (twenty one) (twenty two) In the formula, This represents the total load value during the off-peak period after optimization. This represents the total load value during off-peak hours before optimization. Indicates the number of hours during the off-peak period; This represents the total load value during peak hours after optimization. This represents the total load value during peak hours before optimization. Indicates the number of hours during peak hours.

[0112] The total load value and time period length during high and low periods before and after optimization were both calculated based on the results of typical day division. The results of each indicator after optimization are shown in Table 4.

[0113] Table 4 Results of Optimized Indicators Numerical analysis shows that the time-of-use pricing optimization model proposed in this invention can effectively guide the spatiotemporal transfer of electric vehicle charging load, achieving a total load increase rate of over 36% during off-peak hours and a total load reduction rate of over 19% during peak hours. This verifies that the mechanism can accurately match the characteristics of rural power generation and load, smooth the power grid load curve, improve the local consumption capacity of new energy, and ensure user economic benefits, thus possessing good engineering application value.

[0114] In addition, this invention also proposes a user-side time-of-use electricity pricing optimization device 800.

[0115] Figure 8 A schematic diagram of a user-side time-of-use electricity price optimization device 800 according to an embodiment of the present invention is shown. The user-side time-of-use electricity price optimization device 800 can be deployed in a computing device (such as the aforementioned computing device 100), and the user-side time-of-use electricity price optimization device 800 can be configured to execute the user-side time-of-use electricity price optimization method 200 for rural areas to absorb renewable energy according to the present invention.

[0116] like Figure 8 As shown, in this embodiment of the invention, the user-side time-of-use electricity pricing optimization device 800 includes an acquisition module 810, a determination module 820, a clustering module 830, a time period division module 840, a construction module 850, and a solution module 860, which are coupled in sequence.

[0117] Among them, the acquisition module 810 can acquire multi-dimensional power data in rural areas, including new energy output data, user-side load data, and historical electricity price data.

[0118] The determination module 820 can determine the original daily net load curve based on the user-side load data and the new energy output data.

[0119] The clustering module 830 can cluster the original daily net load curves to obtain typical daily net load curves, which include the net load for each time period of a typical day.

[0120] The time period segmentation module 840 can refine the peak and valley time periods based on the typical daily net load curve to obtain the peak and valley time period segmentation results, which include peak time periods, high-peak time periods, normal time periods, low-peak time periods, and deep valley time periods.

[0121] The construction module 850 can construct a peak-valley difference objective function with the goal of minimizing the peak-valley difference rate of the rural power system. Based on the peak-valley difference objective function and the peak-valley time period division results, a time-of-use pricing optimization model is constructed. The peak-valley difference rate represents the ratio of the difference between the maximum load and the minimum load in each time period after the time-of-use pricing optimization to the maximum load. The load in each time period after the time-of-use pricing optimization can be determined based on the original load and original electricity price of each time period before the time-of-use pricing optimization, the electricity price of each time period after the time-of-use pricing optimization, and the elasticity coefficient of electricity demand in each time period to the change of electricity price in each time period.

[0122] The solver module 860 can solve the time-of-use pricing optimization model to obtain the optimal time-of-use pricing scheme. The optimal time-of-use pricing scheme is used to indicate the optimal peak period price, high-peak period price, normal period price, low-peak period price, and deep-valley period price.

[0123] In some embodiments, the renewable energy output data includes renewable energy generation in each period before the time-of-use pricing optimization, the user-side load data includes the original load in each period before the time-of-use pricing optimization, and the historical electricity price data includes the original electricity price in each period before the time-of-use pricing optimization.

[0124] In some embodiments, the construction module 850 may construct a time-of-use electricity price optimization model based on the objective function of minimizing the peak-valley difference rate and in combination with the peak-valley time period division results in the following manner: constructing an elasticity matrix, which includes the elasticity coefficients of electricity price changes in each time period on the electricity demand in each time period; and constructing a time-of-use electricity price optimization model based on the objective function of minimizing the peak-valley difference rate and the elasticity matrix, in combination with the peak-valley time period division results.

[0125] In some embodiments, the construction module 850 may further construct a time-of-use electricity price optimization model based on the objective function of minimizing the peak-valley difference rate and the elasticity matrix, combined with the peak-valley time period division results, in the following manner: constructing a time-of-use electricity price optimization model based on the objective function of minimizing the peak-valley difference rate, the elasticity matrix, and multiple constraints, combined with the peak-valley time period division results. Several constraints include: rural user electricity cost constraint, indicating that the total electricity cost paid by rural users after time-of-use pricing optimization is not higher than the total electricity cost paid by rural users before time-of-use pricing optimization; peak-valley time-of-use pricing constraint, indicating that the peak-hour price is greater than the normal-hour price, which is greater than the low-hour price, which is greater than the deep-valley price; peak-valley time-of-use load constraint before and after optimization, indicating that the peak-hour load before time-of-use pricing optimization is greater than or equal to the peak-hour load after time-of-use pricing optimization, and that the valley-hour load before time-of-use pricing optimization is less than or equal to the valley-hour load after time-of-use pricing optimization; rural user regulation capacity constraint, indicating that the change in adjustable load of rural users is less than or equal to the proportion of adjustable load; power balance constraint, indicating that the total daily load after time-of-use pricing optimization is equal to the total daily load before time-of-use pricing optimization; and renewable energy consumption constraint, indicating that the proportion of renewable energy generation in rural areas to the total electricity consumption of the power system is greater than or equal to the minimum renewable energy consumption ratio.

[0126] In some embodiments, the net load for each time period in the original daily net load curve is taken as a sample point. Figure 9 A schematic diagram of a clustering module 830 according to some embodiments of the present invention is shown.

[0127] like Figure 9 As shown, the clustering module 830 includes a traversal unit 831, a first selection unit 832, a distance calculation unit 833, a second selection unit 834, a forming unit 835, a profile coefficient calculation unit 836, an average profile coefficient calculation unit 837, a determination unit 838, and a clustering unit 839, which are coupled in sequence.

[0128] The traversal unit 831 can traverse each candidate K value within a predetermined search range. For each candidate K value: The first selection unit 832 can randomly select a sample point from the original daily net load curve as the first cluster center.

[0129] The distance calculation unit 833 can calculate the shortest distance between each sample point in the original daily net load curve and the first cluster center, and determine the probability that the sample point will be selected as the next cluster center based on the shortest distance.

[0130] The second selection unit 834 can select the next cluster center based on the probability that each sample point is selected as the next cluster center, until K cluster centers are obtained.

[0131] Unit 835 can utilize the K-means algorithm to form K clusters based on the original daily net load curve and K cluster centers.

[0132] The profile coefficient calculation unit 836 can calculate the profile coefficient of each sample point in the original daily net load curve based on the average distance between the sample point and other sample points in its cluster, and the average distance between the sample point and all points in its nearest neighbor cluster.

[0133] The average profile coefficient calculation unit 837 can calculate the average profile coefficient of all sample points in the original daily net load curve based on the profile coefficient of each sample point, and use it as the average profile coefficient corresponding to the candidate K value.

[0134] The determining unit 838 can determine the maximum average profile coefficient among the average profile coefficients corresponding to each candidate K value, and determine the candidate K value corresponding to the maximum average profile coefficient as the optimal K value.

[0135] Clustering unit 839 can take the optimal K value and the cluster centers under the optimal K value as the final clustering result, and obtain the typical daily net load curve based on the final clustering result.

[0136] In some embodiments, the forming unit 835 may utilize the K-means algorithm to form K clusters based on the original daily net load curve and K cluster centers according to the following steps: For each sample point in the original daily net load curve, the sample point is assigned to the cluster center closest to the sample point in the K cluster centers to form K clusters; a new cluster center for each cluster is determined, specifically, the mean of all sample points in each cluster is calculated as the new cluster center for each cluster; the above steps are repeated until the cluster center of each cluster no longer changes or the maximum number of iterations is reached, to obtain the final K cluster centers and K clusters.

[0137] In some embodiments, the time period segmentation module 840 can refine the peak and valley time periods based on the typical daily net load curve in the following way: based on the typical daily net load curve, a minimum time period segmentation constraint factor is set and the minimum DBI value is used as the objective function to construct a peak and valley time period segmentation model; and the peak and valley time periods are refined based on the peak and valley time period segmentation model.

[0138] In some embodiments, the solving module 860 may solve the time-of-use electricity pricing optimization model according to the following steps to obtain the optimal time-of-use electricity pricing scheme: Initialize the algorithm parameters, which include particle swarm size, maximum particle velocity, minimum particle velocity, and maximum number of iterations. The particle positions of each particle in the particle swarm are used to represent the time-of-use electricity pricing scheme.

[0139] Based on the algorithm parameters and peak-valley electricity price constraints, the initial particle position and initial particle velocity of each particle in the particle swarm are randomly generated.

[0140] Using the objective function of minimizing peak-to-valley difference, the fitness value of each particle in the particle swarm is calculated based on the particle position of each particle in the particle swarm.

[0141] Based on the fitness value of each particle in the swarm, the swarm is divided into an exploration subgroup and an exploit subgroup. The exploration subgroup and the exploit subgroup can respectively adopt a global search strategy and a local exploit strategy to update the particle position and particle velocity.

[0142] The above steps are executed iteratively, and random complementary solutions are introduced during the iteration process.

[0143] Each iteration executes a predetermined number of times, and particles are migrated between the exploration and development subgroups based on the fitness improvement rate of each particle in the particle swarm to reorganize the particle swarm.

[0144] When the termination condition is met, the optimal particle position in the particle swarm is taken as the optimal time-of-use electricity price scheme. The termination condition includes reaching the maximum number of iterations or the particle positions no longer changing.

[0145] It should be understood that the acquisition module 810, determination module 820, clustering module 830, time period segmentation module 840, construction module 850, and solution module 860 are respectively used to execute the aforementioned steps 210-260. Here, the specific execution logic of each module can be found in the description of steps 210-260 in method 200 above, and will not be repeated here.

[0146] According to the user-side time-of-use pricing optimization method 200 and user-side time-of-use pricing optimization device 800 for new energy consumption in rural areas according to embodiments of the present invention, the original daily net load curves of rural areas are clustered using an improved K-means++ algorithm to obtain typical daily net load curves. Based on the typical daily net load curves, the peak-valley time periods are further subdivided into five types: peak period, high-peak period, normal period, low-valley period, and deep-valley period. Then, the objective function of minimizing the peak-valley difference rate of the rural power system is constructed with the goal of minimizing the peak-valley difference rate. The time-of-use pricing optimization model is constructed by combining the peak-valley time period division results. Finally, the optimal time-of-use pricing scheme can be obtained by solving the time-of-use pricing optimization model using the RODDPSO algorithm. Based on this, the present invention utilizes an improved K-means++ algorithm to automatically determine the optimal number of clusters, thereby improving the algorithm's convergence speed and clustering quality. It further refines the peak-valley periods based on typical daily net load curves, and on this basis, constructs a time-of-use pricing optimization model that considers the elasticity between rural users' electricity demand and electricity prices. The optimal time-of-use pricing scheme determined in this way can effectively incentivize adjustable loads on the user side to participate in demand response and guide the spatiotemporal transfer of loads, thus accurately matching the spatiotemporal characteristics of rural renewable energy output and load, solving the problem of source-load spatiotemporal mismatch, and smoothing the rural power grid net load curve while improving renewable energy absorption.

[0147] Furthermore, this invention constructs a peak-valley time period segmentation model based on a typical daily net load curve, setting a minimum time period segmentation constraint factor and using the minimum DBI value as the objective function. Based on this model, the peak-valley time periods are further refined. This allows for the dynamic determination of the time boundaries for five types of peak-valley time periods.

[0148] The various techniques described herein can be implemented in combination with hardware or software, or a combination thereof. Thus, the methods and apparatus of the present invention, or certain aspects or portions thereof, can take the form of program code (i.e., instructions) embedded in a tangible medium, such as a removable hard disk, USB flash drive, floppy disk, CD-ROM, or any other machine-readable storage medium, wherein when the program is loaded into and executed by a machine such as a computer, the machine becomes an apparatus for practicing the present invention.

[0149] When the program code is executed on a programmable computer, the mobile terminal generally includes a processor, a processor-readable storage medium (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. The memory is configured to store program code; the processor is configured to execute the user-side time-of-use electricity pricing optimization method for rural renewable energy consumption according to instructions in the program code stored in the memory.

[0150] By way of example, and not limitation, readable media include readable storage media and communication media. Readable storage media stores information such as computer-readable instructions, data structures, program modules, or other data. Communication media generally embodies computer-readable instructions, data structures, program modules, or other data in the form of modulated data signals such as carrier waves or other transmission mechanisms, and includes any information delivery medium. Any combination of the above is also included within the scope of readable media.

[0151] In the specification provided herein, the algorithms and displays are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used with the examples of this invention. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0152] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0153] Similarly, it should be understood that, in order to streamline this disclosure and aid in understanding one or more of the various aspects of the invention, in the description of exemplary embodiments of the invention above, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof.

[0154] Those skilled in the art will understand that modules, units, or components of the devices disclosed in the examples herein can be arranged in the devices described in this embodiment, or alternatively, can be located in one or more devices different from the devices in this example. The modules in the foregoing examples can be combined into a single module or, in addition, can be divided into multiple sub-modules.

[0155] Unless otherwise specified, the use of ordinal numbers such as “first,” “second,” “third,” etc., to describe ordinary objects merely indicates different instances of similar objects and is not intended to imply that the objects being described must have a given order in time, space, ordering, or any other manner.

Claims

1. A user-side time-of-use electricity price optimization method for new energy consumption in rural areas, executed in a computing device, comprising: obtaining multi-dimensional power data of rural areas, the multi-dimensional power data comprising new energy output data, user-side load data, and historical electricity price data; determining an original daily net load curve according to the user-side load data and the new energy output data; clustering the original daily net load curve to obtain a typical daily net load curve, the typical daily net load curve comprising typical daily net loads of each time period; refining and dividing peak-valley time periods based on the typical daily net load curve to obtain a peak-valley time period division result, the peak-valley time period division result comprising a peak time period, a high peak time period, a flat time period, a low valley time period, and a deep valley time period; constructing a minimum peak-valley difference rate objective function with a minimum peak-valley difference rate of the rural area power system as a target, and constructing a time-of-use electricity price optimization model based on the minimum peak-valley difference rate objective function and the peak-valley time period division result, the peak-valley difference rate representing a ratio of a difference between a maximum load and a minimum load in each time period after time-of-use electricity price optimization to the maximum load, wherein the load in each time period after time-of-use electricity price optimization is determined according to original loads and original electricity prices of each time period before time-of-use electricity price optimization, electricity prices of each time period after time-of-use electricity price optimization, and an elasticity coefficient of electricity demand of each time period with respect to a change in electricity price of each time period; solving the time-of-use electricity price optimization model to obtain an optimal time-of-use electricity price scheme, the optimal time-of-use electricity price scheme indicating optimal peak time period electricity prices, high peak time period electricity prices, flat time period electricity prices, low valley time period electricity prices, and deep valley time period electricity prices.

2. The method of claim 1, wherein: the new energy output data comprises new energy generation amounts of each time period before time-of-use electricity price optimization, the user-side load data comprises original loads of each time period before time-of-use electricity price optimization, and the historical electricity price data comprises original electricity prices of each time period before time-of-use electricity price optimization.

3. The method of claim 1 or 2, wherein, constructing the time-of-use electricity price optimization model based on the minimum peak-valley difference rate objective function and the peak-valley time period division result comprises: constructing an elasticity matrix, the elasticity matrix comprising elasticity coefficients of electricity demand of each time period with respect to a change in electricity price of each time period; constructing the time-of-use electricity price optimization model based on the minimum peak-valley difference rate objective function, the elasticity matrix, and the peak-valley time period division result.

4. The method of claim 3, wherein, constructing the time-of-use electricity price optimization model based on the minimum peak-valley difference rate objective function and the elasticity matrix, and the peak-valley time period division result comprises: constructing the time-of-use electricity price optimization model based on the minimum peak-valley difference rate objective function, the elasticity matrix, and a plurality of constraint conditions, and the peak-valley time period division result, wherein the plurality of constraint conditions comprise: a rural user electricity cost constraint condition indicating that a total electricity payment of rural users after time-of-use electricity price optimization is not higher than a total electricity payment of rural users before time-of-use electricity price optimization; a peak-valley time period electricity price constraint condition indicating that a peak time period electricity price is greater than a high peak time period electricity price, which is greater than a flat time period electricity price, which is greater than a low valley time period electricity price, which is greater than a deep valley time period electricity price; and The peak-valley period load constraint condition before and after optimization represents that the peak period load before TOU optimization is greater than or equal to the peak period load after TOU optimization, and the valley period load before TOU optimization is less than or equal to the valley period load after TOU optimization; The rural user adjustment capacity constraint condition represents that the adjustable load change amount of the rural user is less than or equal to the adjustable load proportion amount; The power balance constraint condition represents that the total load after TOU optimization is equal to the total load before TOU optimization; The new energy consumption constraint condition represents that the proportion of new energy generation in the total power consumption of the power system in the rural area is greater than or equal to the minimum new energy consumption proportion.

5. The method of any one of claims 1-4, wherein, Each time period net load in the original daily net load curve is taken as a sample point; The original daily net load curve is clustered to obtain a typical daily net load curve, including: Each candidate K value in a predetermined search range is traversed, and the following steps are performed for each candidate K value: A sample point is randomly selected from the original daily net load curve as a first clustering center; For each sample point in the original daily net load curve, the shortest distance between the sample point and the first clustering center is calculated, and the probability of the sample point being selected as a next clustering center is determined according to the shortest distance; According to the probability of each sample point being selected as a next clustering center, a next clustering center is selected until K clustering centers are obtained; K clusters are formed based on the original daily net load curve and the K clustering centers by using a K-means algorithm; For each sample point in the original daily net load curve, the silhouette coefficient of the sample point is calculated according to the average distance between the sample point and other sample points in the cluster and the average distance between the sample point and all points in the nearest neighbor cluster; According to the silhouette coefficient of each sample point, the average silhouette coefficient of all sample points in the original daily net load curve is calculated as the average silhouette coefficient corresponding to the candidate K value; The maximum average silhouette coefficient in the average silhouette coefficients corresponding to each candidate K value is determined, and the candidate K value corresponding to the maximum average silhouette coefficient is determined as the optimal K value; The optimal K value and each clustering center under the optimal K value are taken as a final clustering result, and a typical daily net load curve is obtained according to the final clustering result.

6. The method of claim 5, wherein, K clusters are formed based on the original daily net load curve and the K clustering centers by using a K-means algorithm, including the steps of: For each sample point in the original daily net load curve, the sample point is assigned to the nearest clustering center among the K clustering centers to form K clusters; The new clustering center of each cluster is re-determined, including calculating the mean value of all sample points in each cluster as the new clustering center of each cluster; The above steps are repeatedly performed until the clustering center of each cluster no longer changes or the maximum iteration number is reached, to obtain the final K clustering centers and K clusters.

7. The method of any one of claims 1-6, wherein, The peak-valley period refinement division is performed based on the typical daily net load curve, including: Based on the typical daily net load curve, a minimum time segment division constraint factor is set and a DBI value is taken as an objective function to construct a peak-valley time segment division model; Based on the peak-valley time segment division model, a peak-valley time segment refinement division is performed.

8. The method of any one of claims 1-7, wherein, The time-of-use electricity price optimization model is solved to obtain an optimal time-of-use electricity price scheme, including the steps of: Initializing algorithm parameters, the algorithm parameters including particle swarm size, maximum particle speed, minimum particle speed, and maximum iteration number, and a particle position of each particle in the particle swarm is used to represent a time-of-use electricity price scheme; Based on the algorithm parameters and the peak-valley time segment electricity price constraint condition, an initial particle position and an initial particle speed of each particle in the particle swarm are randomly generated; Based on the particle position of each particle in the particle swarm, the fitness value of each particle in the particle swarm is calculated by using a peak-valley difference rate minimum objective function; According to the fitness value of each particle in the particle swarm, the particle swarm is divided into an exploration subgroup and a development subgroup, and the exploration subgroup and the development subgroup are respectively adapted to update the particle position and the particle speed by using a global search strategy and a local development strategy; The above steps are iteratively performed, and a random opposite solution is introduced in the iteration process; Every iteration is performed for a predetermined number of times, and particle migration is performed between the exploration subgroup and the development subgroup according to the fitness improvement rate of each particle in the particle swarm to recombine the particle swarm; When a termination condition is met, the optimal particle position in the particle swarm is taken as an optimal time-of-use electricity price scheme, wherein the termination condition is met when the maximum iteration number is reached or the particle position no longer changes.

9. A user-side time-of-use electricity price optimization device deployed in a computing device, comprising: An acquisition module adapted to acquire multi-dimensional power data of a rural area, the multi-dimensional power data including new energy output data, user-side load data, and historical electricity price data; A determination module adapted to determine an original daily net load curve according to the user-side load data and the new energy output data; A clustering module adapted to cluster the original daily net load curve to obtain a typical daily net load curve, the typical daily net load curve including a typical daily net load of each time segment; A time segment division module adapted to perform peak-valley time segment refinement division based on the typical daily net load curve to obtain a peak-valley time segment division result, the peak-valley time segment division result including a sharp peak time segment, a high peak time segment, a flat time segment, a low valley time segment, and a deep valley time segment; A construction module adapted to construct a peak-valley difference rate minimum objective function with a peak-valley difference rate minimum as an objective, and construct a time-of-use electricity price optimization model based on the peak-valley difference rate minimum objective function and in combination with the peak-valley time segment division result, the peak-valley difference rate representing a ratio of a difference between a maximum load and a minimum load of each time segment after time-of-use electricity price optimization to the maximum load, wherein the maximum load and the minimum load of each time segment after time-of-use electricity price optimization are determined according to an original load and an original electricity price of each time segment before time-of-use electricity price optimization, a time-of-use electricity price of each time segment after time-of-use electricity price optimization, and an elasticity coefficient of each time segment electricity price change to each time segment electricity demand. The solving module is adapted to solve the time-of-use electricity price optimization model to obtain an optimal time-of-use electricity price scheme, which is used to indicate an optimal peak time period electricity price, an optimal high peak time period electricity price, an optimal flat time period electricity price, an optimal low valley time period electricity price, and an optimal deep valley time period electricity price.

10. The apparatus of claim 9, wherein, The new energy output data comprises new energy power generation in each time period before time-of-use electricity price optimization, the user side load data comprises original load in each time period before time-of-use electricity price optimization, and the historical electricity price data comprises original electricity price in each time period before time-of-use electricity price optimization.

11. The apparatus of claim 9 or 10, wherein, The constructing module is adapted to construct the time-of-use electricity price optimization model based on the minimum peak-valley difference rate objective function and the elasticity matrix in combination with the peak-valley time period division result in the following manner: An elasticity matrix is constructed, and the elasticity matrix comprises elasticity coefficients of electricity price change in each time period to electricity demand in each time period; The time-of-use electricity price optimization model is constructed based on the minimum peak-valley difference rate objective function and the elasticity matrix in combination with the peak-valley time period division result.

12. The apparatus of claim 11, wherein, The constructing module is further adapted to construct the time-of-use electricity price optimization model based on the minimum peak-valley difference rate objective function and the elasticity matrix in combination with the peak-valley time period division result in the following manner: The time-of-use electricity price optimization model is constructed based on the minimum peak-valley difference rate objective function, the elasticity matrix, and a plurality of constraint conditions in combination with the peak-valley time period division result, wherein the plurality of constraint conditions comprise: a rural user electricity cost constraint condition, which indicates that total electricity fees paid by the rural user after time-of-use electricity price optimization is not higher than total electricity fees paid by the rural user before time-of-use electricity price optimization; a peak-valley time period electricity price constraint condition, which indicates that the peak time period electricity price is greater than the high peak time period electricity price, which is greater than the flat time period electricity price, which is greater than the low valley time period electricity price, which is greater than the deep valley time period electricity price; a before-and-after optimization peak-valley time period load constraint condition, which indicates that the peak time period load before time-of-use electricity price optimization is greater than or equal to the peak time period load after time-of-use electricity price optimization, and the valley time period load before time-of-use electricity price optimization is less than or equal to the valley time period load after time-of-use electricity price optimization; a rural user adjustment capability constraint condition, which indicates that an adjustable load change amount of the rural user is less than or equal to an adjustable load proportion amount; an electricity amount balance constraint condition, which indicates that total daily load after time-of-use electricity price optimization is equal to total daily load before time-of-use electricity price optimization; a new energy consumption constraint condition, which indicates that a proportion of new energy power generation in the rural area to total electricity demand of the power system is greater than or equal to a minimum new energy consumption proportion.

13. The apparatus of any of claims 9-12, wherein, Each time period net load in the original daily net load curve is taken as a sample point; The clustering module comprises: a traversal unit adapted to traverse each candidate K value in a predetermined search range; a first selection unit adapted to randomly select a sample point from the original daily net load curve as a first clustering center; a distance calculation unit adapted to calculate, for each sample point in the original daily net load curve, a shortest distance between the sample point and the first clustering center, and determine a probability that the sample point is selected as a next clustering center according to the shortest distance; and a second selection unit adapted to select, according to the probability, a sample point with the highest probability as the next clustering center. The second selection unit is adapted to select the next cluster center according to a probability of each sample point being selected as the next cluster center until K cluster centers are obtained. The forming unit is adapted to form K clusters based on the original daily net load curve and the K cluster centers by using a K-means algorithm. The silhouette coefficient calculation unit is adapted to calculate, for each sample point in the original daily net load curve, a silhouette coefficient of the sample point according to an average distance between the sample point and other sample points in a cluster where the sample point is located and an average distance between the sample point and all points in a nearest neighbor cluster. The average silhouette coefficient calculation unit is adapted to calculate an average silhouette coefficient of all sample points in the original daily net load curve as an average silhouette coefficient corresponding to the candidate K value according to the silhouette coefficient of each sample point. The determination unit is adapted to determine a maximum average silhouette coefficient among the average silhouette coefficients corresponding to the respective candidate K values, and determine the candidate K value corresponding to the maximum average silhouette coefficient as an optimal K value. The clustering unit is adapted to take the optimal K value and the respective cluster centers under the optimal K value as a final clustering result, and obtain a typical daily net load curve according to the final clustering result.

14. The apparatus of claim 13, wherein, The forming unit is adapted to form K clusters based on the original daily net load curve and the K cluster centers by using a K-means algorithm according to the following steps: For each sample point in the original daily net load curve, the sample point is assigned to a cluster center closest to the sample point among the K cluster centers to form K clusters. The new cluster center of each cluster is re-determined, including: calculating a mean value of all sample points in each cluster as a new cluster center of each cluster. The above steps are repeatedly executed until the cluster center of each cluster no longer changes or a maximum iteration number is reached, to obtain the final K cluster centers and K clusters.

15. The apparatus of any one of claims 9-14, wherein, The period division module is adapted to perform peak-valley period refinement division based on the typical daily net load curve according to the following manner: A peak-valley period division model is constructed based on the typical daily net load curve, a minimum period division constraint factor is set, and a minimum DBI value is taken as an objective function. Peak-valley period refinement division is performed based on the peak-valley period division model.

16. The apparatus of any one of claims 9-15, wherein, The solving module is adapted to solve the time-of-use electricity price optimization model to obtain an optimal time-of-use electricity price scheme according to the following steps: An algorithm parameter is initialized, the algorithm parameter including a particle swarm size, a maximum particle speed, a minimum particle speed, and a maximum iteration number, and a particle position of each particle in the particle swarm is used to represent a time-of-use electricity price scheme; An initial particle position and an initial particle speed of each particle in the particle swarm are randomly generated based on the algorithm parameter and a peak-valley period electricity price constraint condition; An adaptability value of each particle in the particle swarm is calculated based on the particle position of each particle in the particle swarm by using a minimum peak-valley difference rate objective function; The particle swarm is divided into an exploration sub-swarm and a development sub-swarm according to the adaptability value of each particle in the particle swarm, and the exploration sub-swarm and the development sub-swarm are respectively adapted to update the particle position and the particle speed by using a global search strategy and a local development strategy. The above steps are iteratively performed, and random opposite solutions are introduced during iteration; The particle migration is performed between the exploration subgroup and the development subgroup according to the fitness improvement rate of each particle in the particle swarm every predetermined number of iterations, so as to recombine the particle swarm; When a termination condition is met, the optimal particle position in the particle swarm is taken as the optimal time-of-use electricity price scheme, wherein the termination condition is met by reaching a maximum number of iterations or the particle position no longer changing.

17. A computing device comprising: at least one processor; and a memory having stored program instructions configured to be processed by the at least one processor, the program instructions comprising instructions for processing the method of any one of claims 1-8.

18. A readable storage medium having stored program instructions, which, when read and executed by a computing device, cause the computing device to perform the method of any one of claims 1-8.

Images