Methods and devices for predicting demand response in power distribution networks

Abstract

This invention provides a method and apparatus for predicting demand response in a distribution network. The method includes: acquiring time-of-use (TOU) compensation prices for demand response in the distribution network and response information declared by power users, the response information including load type, response mode, preferred time, and response power; determining the compensation cost for each power user based on the TOU and response information; determining a multi-dimensional time-series model for each power user based on their compensation cost and response information, the multi-dimensional time-series model characterizing the time-series characteristics of the compensation cost and response power corresponding to each load type; and classifying and aggregating the demand responses of each power user based on their multi-dimensional time-series models to obtain a cluster aggregation model for predicting various types of demand responses in the distribution network. This invention can accurately predict distribution network demand response while reducing the computational load on the distribution network.

Description

Technical Field

[0001] This invention relates to the field of power transmission and transformation technology, and in particular to a method and apparatus for predicting demand response in power distribution networks. Background Technology

[0002] Demand response, as a crucial resource for grid optimization control, can significantly improve the stability of power system operation by mitigating peak-to-valley load differences. Electricity user loads are the primary resource for constructing demand response, and distribution network optimization control encourages load participation. However, the demand-side load response of the distribution network is highly randomized due to changes in compensation benefits, weather conditions, and unforeseen events, resulting in persistently high response errors. Furthermore, demand-side resources face challenges such as a large number of participating loads, their dispersed distribution, and low individual load response capacity, leading to a surge in computational load for grid optimization control and a prolonged optimization time. Summary of the Invention

[0003] This invention provides a method and apparatus for predicting the demand response of a distribution network, which can accurately predict the demand response of the distribution network while reducing the computational load of the distribution network.

[0004] In a first aspect, the present invention provides a method for predicting demand response in a distribution network, comprising: acquiring time-of-use compensation prices for demand response in the distribution network and response information declared by power users, wherein the response information includes load type, response method, preferred time, and response power; determining the compensation cost for each power user based on the time-of-use compensation prices and response information; determining a multi-dimensional time-series model for each power user based on the compensation cost and response information, wherein the multi-dimensional time-series model is used to characterize the time-series characteristics of the compensation cost and the time-series characteristics of the response power corresponding to each load type; and classifying and aggregating the demand responses of each power user based on the multi-dimensional time-series model to obtain a cluster aggregation model for predicting various types of demand responses in the distribution network.

[0005] This invention provides a method for predicting demand response in a distribution network. By calculating the compensation costs of each power user and constructing a multi-dimensional time-series model for each power user, the method characterizes the time-series characteristics of the compensation costs and response power corresponding to each load type. Thus, this invention can use the multi-dimensional time-series model to characterize the changes in load type, compensation costs, and response power of each power user's demand response over time, thereby classifying and aggregating the demand responses of each power user. Based on the aggregated cluster model, the method predicts various types of demand responses in the distribution network, achieving accurate prediction of distribution network demand response and reducing the computational load of the distribution network.

[0006] In one possible implementation, the unit capacity compensation cost for each power user is determined based on the time-of-use compensation price and response information. This includes: determining the unit capacity compensation cost based on the time-of-use compensation price and a preset response mode coefficient, where the response mode coefficient is positively correlated with the flexibility of demand response; determining the unit time compensation cost based on the time-of-use compensation price, preferred time, and expected response time; and determining the compensation cost for each power user based on the unit capacity compensation cost and the unit time compensation cost.

[0007] In one possible implementation, a multi-dimensional time-series model for each power user is determined based on their compensation costs and response information. This includes: for any power user, determining the change of compensation costs over time based on the user's compensation costs; determining the change of response power over time for each load type based on the user's load type, response method, preferred time, and response power; and determining the multi-dimensional time-series model for the power user based on the changes in compensation costs and response power over time for each load type. The multi-dimensional time-series model for the power user represents the time series using a binary tree hierarchical structure, the response power using node functions of the binary tree, and the compensation cost using inter-node functions of the binary tree.

[0008] In one possible implementation, based on the multi-dimensional time-series models of each power user, the demand responses of each power user are classified and aggregated to obtain a cluster aggregation model. This includes: using load type, compensation cost, and response power in the multi-dimensional time-series models as feature variables, classifying and aggregating to obtain sub-load clusters, each sub-load cluster representing the sum of response power of multiple power users in a type of demand response; and based on each sub-load cluster, a cluster aggregation model is obtained, which includes load reduction clusters, load shifting clusters, and load transfer clusters, wherein each load cluster includes multiple sub-load clusters.

[0009] In one possible implementation, after classifying and aggregating the demand responses of each power user based on a multi-dimensional time-series model to obtain a cluster aggregation model, the following steps are also included: calculating the load response rate corresponding to each load type based on each sub-load cluster; calculating the uncertainty coefficient corresponding to each load type based on the load response rate corresponding to each load type; the uncertainty coefficient is used to characterize the magnitude of the response power deviation caused by the uncertainty of the demand response; and correcting each sub-load cluster based on the uncertainty coefficient corresponding to each load type to obtain the corrected sub-load clusters.

[0010] Secondly, embodiments of the present invention provide a control device for a distribution network, comprising: a communication module for acquiring time-of-use compensation tariffs for demand response in the distribution network and response information declared by power users, the response information including load type, response method, preferred time, and response power; a processing module for determining the compensation cost for each power user based on the time-of-use compensation tariffs and response information; determining a multi-dimensional time-series model for each power user based on the compensation cost and response information, the multi-dimensional time-series model being used to characterize the time-series characteristics of the compensation cost and response power corresponding to each load type; and classifying and aggregating the demand responses of each power user based on the multi-dimensional time-series model to obtain a cluster aggregation model for predicting various types of demand responses in the distribution network.

[0011] In one possible implementation, the processing module is specifically used to determine the unit capacity compensation cost based on the time-of-use compensation price and a preset response mode coefficient, wherein the response mode coefficient is positively correlated with the flexibility of demand response; to determine the unit time compensation cost based on the time-of-use compensation price, preferred time, and expected response time; and to determine the compensation cost for each electricity user based on the unit capacity compensation cost and the unit time compensation cost.

[0012] In one possible implementation, the processing module is specifically used to, for any power user, determine the change of compensation cost over time based on the power user's compensation cost; determine the change of response power over time for each load type based on the power user's load type, response method, preferred time, and response power; and determine the multi-dimensional time series model of the power user based on the change of compensation cost over time and the change of response power over time for each load type. The multi-dimensional time series model of the power user represents the time series using a binary tree hierarchical structure, the response power using node functions of the binary tree, and the compensation cost using inter-node functions of the binary tree.

[0013] In one possible implementation, the processing module is specifically used to classify and aggregate load types, compensation costs, and response power in a multi-dimensional time series model as feature variables to obtain sub-load clusters. Each sub-load cluster represents the sum of response power of multiple electricity users in a type of demand response. Based on each sub-load cluster, a cluster aggregation model is obtained. The cluster aggregation model includes load reduction clusters, load shifting clusters, and load transfer clusters, wherein each load cluster includes multiple sub-load clusters.

[0014] In one possible implementation, the processing module is further configured to calculate the load response rate corresponding to each load type based on each sub-load cluster; calculate the uncertainty coefficient corresponding to each load type based on the load response rate corresponding to each load type; the uncertainty coefficient is used to characterize the magnitude of the response power deviation caused by the uncertainty of the demand response; and correct each sub-load cluster based on the uncertainty coefficient corresponding to each load type to obtain the corrected sub-load clusters.

[0015] Thirdly, embodiments of the present invention provide an electronic device, characterized in that the electronic device includes a memory and a processor, the memory storing a computer program, and the processor being configured to call and run the computer program stored in the memory to perform the steps of the method as described in the first aspect and any possible implementation thereof.

[0016] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the method as described in the first aspect and any possible implementation thereof.

[0017] The technical effects of any of the implementation methods in the second to fourth aspects mentioned above can be found in the technical effects of the corresponding implementation method in the first aspect, and will not be repeated here. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a distribution network demand response prediction method provided in an embodiment of the present invention;

[0020] Figure 2 This is a schematic diagram of a multi-dimensional time-series model of a power user provided in an embodiment of the present invention;

[0021] Figure 3 This is a schematic diagram of another multi-dimensional time-series model of a power user provided in an embodiment of the present invention;

[0022] Figure 4 This is a schematic diagram of another multi-dimensional time-series model of a power user provided in an embodiment of the present invention;

[0023] Figure 5This is a schematic diagram of a cluster aggregation model provided in an embodiment of the present invention;

[0024] Figure 6 This is a schematic diagram illustrating the variation of load response rate with compensation electricity price according to an embodiment of the present invention;

[0025] Figure 7 This is a schematic diagram of the structure of a power distribution network control device provided in an embodiment of the present invention;

[0026] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0027] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0028] In the description of this invention, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" and "more than one" refer to two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0029] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner to facilitate understanding.

[0030] Furthermore, the terms "comprising" and "having," and any variations thereof, used in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not limited to the steps or modules listed, but may optionally include other steps or modules not listed, or may optionally include other steps or modules inherent to such process, method, product, or device.

[0031] To make the objectives, technical solutions, and advantages of the present invention clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0032] As described in the background section, with the development of distribution network optimization, the proportion of demand response in the distribution network is gradually increasing, leading to a decrease in the effectiveness of distribution network optimization control. Currently, there is a demand response forecasting scheme that establishes a load response model considering load response status and load operating time, alleviating the problem of describing load response time-series power. However, this scheme does not consider the time-series cost of load response and the uncertainties caused by clustering, resulting in a significant increase in the computational load of power grid optimization control and persistently high response errors.

[0033] To solve the above technical problems, such as Figure 1 As shown, this embodiment of the invention provides a distribution network demand response prediction method, the execution subject of which is the control device of the distribution network, and the distribution network demand response prediction method includes steps S101-S104.

[0034] S101. Obtain the time-of-use compensation price for demand response in the distribution network and the response information declared by electricity users themselves.

[0035] In this embodiment of the application, the response information includes load type, response method, preferred time and response power.

[0036] In some embodiments, the response information may also include response capacity, response period, and response frequency.

[0037] In some embodiments, load types may include loads that can be reduced, loads that can be moved, and loads that can be transferred.

[0038] For example, as shown in the table, the unit compensation electricity price for different types of loads at different times is as follows. Taking the load that can be reduced as an example, the unit compensation electricity price is 0.2 yuan / kWh from 0:00 to 7:00, 0.3 yuan / kWh from 8:00 to 10:00, 0.5 yuan / kWh from 11:00 to 14:00, 0.4 yuan / kWh from 15:00 to 18:00, 0.7 yuan / kWh from 19:00 to 21:00, and 0.4 yuan / kWh from 22:00 to 24:00.

[0039] Table 1

[0040]

[0041] It should be noted that the response information is submitted by the electricity user himself. The submission process is based on the following assumptions: 1) Users can decide whether to participate in the demand response the next day before the deadline. If they participate, they should submit the response information to the distribution network for the next day; 2) The response mode of the user is identified by 0-1, where P1 indicates that the response can be continuous or distributed, and P0 indicates that the response mode of the load is only continuous. Users need to explain the load response mode when submitting the information.

[0042] For example, as shown in Table 2, this embodiment presents the response information submitted by a power user. Taking the load that can be reduced as an example, the response power of the load is 1KW, the response capacity is 2KWh, the response period is 11:00-13:00, the preferred response time is 11:30, the response frequency is 2, and the response interval is 0.25h.

[0043] Table 2

[0044]

[0045] S102. Based on the time-of-use compensation price and response information, determine the compensation cost for each electricity user.

[0046] It should be noted that the unit time-series compensation cost of the load is affected by user response characteristics, mainly reflected in response methods and time deviations. Response methods can be divided into continuous response and distributed response. Since the two response methods differ significantly in response flexibility, a response method coefficient is proposed to provide additional compensation for loads with higher flexibility. Time deviation refers to the fact that users have the strongest willingness to respond and the highest initiative in participating in dispatch when responding to dispatch instructions within the expected time. As the response time deviates from the expected response time, the user's willingness to respond decreases; therefore, additional electricity price compensation is provided to users to enhance their willingness to respond.

[0047] As one possible implementation, the control device can determine the compensation cost for each power user based on steps S1021-S1023.

[0048] S1021. Determine the unit capacity compensation cost based on the time-of-use compensation price and the preset response mode coefficient.

[0049] In some embodiments, the response mode coefficient is positively correlated with the flexibility of demand response.

[0050] For example, when the load type is continuous response, the load response coefficient ξ i The load response coefficient ξ is 1 when the load type is distributed response. i It is 1.1.

[0051] For example, such as Figure 2As shown, assuming the load type is a reducible load, the expected response time is 11:30, the actual response time is 11:45, and the response method is distributed response, this invention can calculate the response method compensation cost, the response time offset compensation cost, and further calculate the user timing compensation cost.

[0052] S1022. Determine the compensation cost per unit time based on the time-of-use compensation price, preferred time, and expected response time.

[0053] S1023. Determine the compensation cost for each electricity user based on the unit capacity compensation cost and the unit time compensation cost.

[0054] For example, the control device can determine the compensation cost for the electricity user based on the following formula.

[0055]

[0056] Among them, c i (t) represents the unit capacity time-series compensation cost, which is the sum of capacity compensation cost and time compensation cost; The response capacity compensation cost for load i at time t; ξ represents the time compensation cost of load i at time t; c(t) represents the unit capacity compensation price published by the distribution network; i t is the response mode coefficient for load i; i t represents the actual response time. i,0 Let t be the expected response time, and t s,i ≤t i ≤t e,i , t s,i The minimum response time reported by the user, t e,i The maximum response time reported by the user.

[0057] It should be noted that this cost model for compensating user responses can simultaneously take into account the impact of response method and preferred time on compensation costs. When responses are distributed, load response flexibility is high, but compensation costs are low otherwise. When the actual response time deviation is small, time compensation costs are low and response willingness is high, so low compensation can achieve an incentive effect. When the actual response time deviation is large, time compensation costs surge, response willingness is low, and compensation grows quadratically, effectively stimulating users' initiative in participating in scheduling.

[0058] S103. Based on the compensation costs and response information of each power user, determine the multi-dimensional time series model of each power user.

[0059] In this embodiment, the multi-dimensional time series model is used to characterize the time series characteristics of compensation cost and response power corresponding to each load type.

[0060] As one possible implementation, for any power user, the control device can determine the multi-dimensional time series model of each power user based on steps S1031-S1033.

[0061] S1031. Based on the compensation cost of the electricity user, determine how the compensation cost changes over time.

[0062] S1032. Based on the load type, response mode, preferred time and response power of the power user, determine the change of response power of each load type over time.

[0063] It should be noted that the load response power at any given time exhibits a coupling characteristic with the response power at previous and subsequent times. The temporal coupling characteristic of the response power is mainly affected by the inherent characteristics of the load and user reporting preferences. Therefore, the temporal characteristics of various types of loads differ significantly, and constraints for the temporal characteristics of three types of loads are established respectively.

[0064] In some embodiments, the requirement for load reduction is that the resulting load curve is lower than the target load curve. The load response timing characteristic constraints include reduction period constraints, reduction power constraints, reduction capacity constraints, reduction frequency constraints, and timing compensation cost constraints.

[0065] For example, the timing constraint for reducing load response power can be expressed as the following formula.

[0066]

[0067] in, Let be the response power of the load that can be reduced for load i, where the loads that can be reduced include [1, N1], and N1 is the load number. For load i, the maximum reduction capacity, The power has been reduced at time t. P represents the amount of electricity reduced at time t-1, Δt is the duration between time t-1 and time t, and P is the energy level that has decreased. i const The rated power of load i; load reduction begins at time j, and power is restored at time k, T min,i N is the minimum time interval between two load reductions of load i; i t represents the maximum number of times load i can be reduced during the reporting period. s,i The minimum response time reported by the user, t e,i This represents the maximum response time reported by the user. The response state of load i at time t is represented by "1" indicating that it is in the response state, otherwise it is "0".

[0068] In some embodiments, the requirement for a transferable load is that the resulting load curve after transfer matches the target load curve. The time-series constraints on the response power of a transferable load include transfer time constraints, transfer power constraints, and time-series compensation cost constraints.

[0069] For example, the timing constraint of the shiftable load response power can be expressed as the following formula.

[0070]

[0071] in, Let t be the response power of the transferable load i, where the transferable loads include [N1+1, N2], N1 and N2 are load numbers, and t is the response power of the load i. i For the start-up time of the transferable load i, t s,i The minimum response time reported by the user, t e,i This represents the maximum response time reported by the user.

[0072] In some embodiments, the requirement for transferable loads is that the operating time of the load after transfer is the rated operating time. The timing characteristics constraints of the transferable load response power include transfer period constraints, transfer capacity constraints, uptime constraints, and timing compensation cost constraints.

[0073] For example, the timing constraint of the shiftable load response power can be expressed as the following formula.

[0074]

[0075] in, Let N be the response power of the transferable load i. The transferable loads include [N2+1, N3], where N2 and N3 are load numbers. This represents the maximum transfer capacity of load i; The amount of electricity transferred at time t. The transferred load is E at time t-1. The initial transferable load for user i is 0. Δt is the duration between time t-1 and time t. i The load run time specified for user i. This represents the operating state of load i at time t, where "1" indicates it is in the operating state and "0" indicates it is not. es,i t represents the start time of normal load operation. ee,i For the normal operation and termination time of the load, I i,t Let be the current of load i at time t.

[0076] S1033. Based on the changes in compensation costs over time and the changes in response power for each load type over time, determine the multi-dimensional time-series model for the power user.

[0077] The multi-dimensional time series model of this power user uses a binary tree link hierarchy to represent the time series, a binary tree node function to represent the response power, and a binary tree inter-node function to represent the compensation cost.

[0078] It should be noted that a binary tree, as a non-linear data structure, is a hierarchical set composed of n (n≥0) finite nodes. Its unique hierarchical relationship can describe the temporal characteristics of compensation cost and response power. Therefore, a binary tree model is established based on the constraints of the load response power temporal characteristics, and a multi-dimensional response model of the equipment is established by combining it with the load temporal cost model. The model uses a binary tree link hierarchy to represent the time series, node functions represent response power and load status, and inter-node functions represent the unit capacity compensation cost at that moment.

[0079] For example, based on the user response information in Table 2 and the time-series power constraints of various types of loads, a multi-dimensional time-series model of power user response load can be obtained by combining a binary tree. Figure 2 , Figure 3 , Figure 4 A schematic diagram of a multi-dimensional time-series model for electricity users. Figure 2 , Figure 3 , Figure 4 It can accurately locate the response power and compensation cost of the load at different time periods within the response cycle.

[0080] S104. Based on the multi-dimensional time series model of each power user, the demand response of each power user is classified and aggregated to obtain a cluster aggregation model, so as to predict various types of demand responses in the distribution network.

[0081] As one possible implementation, the control device can classify and aggregate load types, compensation costs, and response power from a multi-dimensional time-series model as feature variables to obtain sub-load clusters. Each sub-load cluster represents the sum of the response power of multiple electricity users in a certain type of demand response. Based on each sub-load cluster, a cluster aggregation model is obtained.

[0082] The cluster aggregation model includes load reduction clusters, load shifting clusters, and load transfer clusters. Each load cluster contains multiple sub-load clusters.

[0083] It should be noted that in actual systems, the demand-side user base is large and response resources are abundant. However, individual users have low load response levels and varying response characteristics. If each user directly participates in demand response, it will significantly increase cloud computing load and increase line communication pressure. If optimization is performed on the overall load of the same type, the specific characteristics of the load will not be taken into account, thereby reducing the optimization effect.

[0084] This application's embodiments generate user sub-load clusters based on load type, user response method, and expected response time as characteristic variables, and establish a distribution area load hierarchical evaluation model based on user cluster division. Users within each sub-load cluster are aggregated to achieve differentiated aggregation models between sub-load clusters and unified aggregation models within sub-load clusters.

[0085] like Figure 5 As shown in the figure, this application embodiment provides a schematic diagram of a cluster aggregation model. Figure 5 The first layer of the cluster aggregation model is the aggregated load, which is the sum of all user loads of the same type. For example, the present invention can sum the response power according to the load type to obtain load reduction clusters, load shifting clusters, and load transfer clusters. Figure 5 The second layer of the cluster aggregation model is the sub-aggregated load, which is based on the result of load cluster partitioning. It is assumed that there are a total of V sub-aggregated loads, and each sub-aggregated load represents the sum of the loads corresponding to that sub-load cluster. Figure 5 The third layer of the cluster aggregation model consists of the individual loads of each power user.

[0086] For example, the cluster aggregation model can be represented by the following formula.

[0087]

[0088] In the formula: P IL (t) represents the response power of the load cluster that can be reduced at time t, P SL (t) represents the response power of the shiftable load cluster at time t, P TL (t) represents the response power of the transferable load cluster at time t; C IL (t) represents the unit capacity compensation cost of the load cluster that can be reduced at time t, C SL (t) represents the unit capacity compensation cost of the shiftable load cluster at time t, C TL (t) represents the unit capacity compensation cost of the transferable load cluster at time t, V IL To reduce the number of sub-load clusters in a load cluster, V SL V represents the number of sub-load clusters of a shiftable load cluster. TL A represents the number of sub-load clusters of the transferable load cluster. k For the kth type of sub-load cluster, Δt is the prediction period of 15 min. The response power of load i that can be reduced. Let be the response power of the transferable load i. Let be the response power of the transferable load i. Let i be the unit capacity compensation cost for load i at time t. Let i be the unit capacity compensation cost of the load that can be shifted at time t. Let $t$ be the unit capacity compensation cost of load $i$ that can be transferred at time $t$.

[0089] This invention provides a method for predicting demand response in a distribution network. By calculating the compensation costs of each power user and constructing a multi-dimensional time-series model for each power user, the method characterizes the time-series characteristics of the compensation costs and response power corresponding to each load type. Thus, this invention can use the multi-dimensional time-series model to characterize the changes in load type, compensation costs, and response power of each power user's demand response over time, thereby classifying and aggregating the demand responses of each power user. Based on the aggregated cluster model, the method predicts various types of demand responses in the distribution network, achieving accurate prediction of distribution network demand response and reducing the computational load of the distribution network.

[0090] It should be noted that the embodiments of this application can establish a three-dimensional spatial coordinate system with time, sub-load cluster unit capacity compensation cost, and sub-load cluster response power as axes. The three-dimensional coordinate system represents the response capabilities of load clusters of the same type within a scheduling week. The classification and aggregation method adopted in the embodiments of this application aggregates users with similar response characteristics and performs unified optimization after aggregation, avoiding problems such as a surge in computational load and increased communication pressure caused by centralized control of all users in the cloud.

[0091] It should be noted that cluster response is affected by many factors such as compensation benefits, psychological changes, and unexpected events. These factors are random and ambiguous, resulting in a high degree of uncertainty in user response behavior.

[0092] This invention establishes a user response information reporting mechanism to obtain demand-side response information. From the perspective of incentive-based demand response, it constructs a multi-dimensional time-series representation model of response load that considers compensation costs and response power. Based on this, a load hierarchical aggregation model is established, and the uncertainty of cluster demand response is calculated using membership degree fuzzy expectation constraints, thereby completing the assessment of the time-series response capability of the transformer area load cluster.

[0093] Optionally, after step S104, the distribution network demand response prediction method provided in this embodiment of the invention further includes steps S201-S203.

[0094] S201. Based on each sub-load cluster, calculate the load response rate corresponding to each load type.

[0095] In some embodiments, such as Figure 6 As shown, the embodiment of the present invention provides the load response rate λ. DR With compensation electricity price C DR The changing pattern. Figure 6The black curve represents the expected load response rate. Curve ① represents the upper limit of response rate fluctuation, and curves ② represent the lower limit of response rate fluctuation. The shaded area represents the range of load response rate fluctuation.

[0096] For example, the control device can determine the load response rate for each load type using the following formula.

[0097]

[0098] In the formula: P IL (t) represents the response power of the load cluster that can be reduced at time t, P SL (t) represents the response power of the shiftable load cluster at time t, P TL (t) represents the response power of the transferable load cluster at time t; λ IL To reduce the load response rate of the load cluster, λ SL λ represents the load response rate of the shiftable load cluster. TL Let P(t) be the load response rate of the transferable load cluster, and let P(t) be the load power at time t.

[0099] S202. Based on the load response rate corresponding to each load type, calculate the uncertainty coefficient corresponding to each load type.

[0100] The uncertainty coefficient is used to characterize the magnitude of the response power deviation caused by the uncertainty of the demand response.

[0101] like Figure 6 As shown, λ DR +d t λ represents the upper limit of the actual load response rate under a certain compensated electricity price. DR -d t This represents the lower limit of the actual load response rate under a given compensation electricity price.

[0102] When the time-of-use compensation electricity price is C DR,0 When users begin to participate in grid response; when the compensation price is at (C DR,0 C DR,max When the load response power is between C and C's, the user response power increases with the increase of the compensation electricity price, and the load response uncertainty decreases accordingly; conversely, the influence of external environmental factors on the load response power will become increasingly greater, and the load response uncertainty will increase. When C's... DR,max At this point, the user response power reaches its maximum expected value, and the fluctuation range of the response power is approximately zero.

[0103] This application uses a triangular membership function to describe the uncertainty of the load response rate. The fuzzy expression for the load response rate under time-of-use pricing is shown below.

[0104]

[0105] in, Let λ be the fuzzy expression for the load response rate at time t; t The load response rate at time t can be calculated using the above formula; For time-of-use compensation pricing; c DR,0 c is the compensation electricity price corresponding to the dead zone inflection point. DR,max λ is the compensation electricity price corresponding to the inflection point of the saturation region. max To compensate for the upper limit of load response rate under the electricity price; d t d represents the maximum deviation of the load response rate. t ≥0.

[0106] Based on the variation pattern of deviations caused by the uncertainty of demand response, the mechanism of the change in the maximum error level of load response rate is expressed by the following formula.

[0107]

[0108] Where, d t This represents the maximum deviation of the load response rate. To compensate for the electricity price, k1 is the ratio of the maximum error level to the load response change rate before the electricity price factor becomes dominant, and k2 is the ratio of the maximum error level to the load response change rate after the electricity price factor becomes dominant. The electricity price is used to compensate for the inflection point. Once the rate of change in electricity prices exceeds the rate of change at the inflection point, electricity price factors begin to play a dominant role, and their absolute value is negatively correlated with the degree of guidance provided by the level of electricity price compensation.

[0109] S203. Based on the uncertainty coefficients corresponding to each load type, each sub-load cluster is corrected to obtain the corrected sub-load clusters.

[0110] The actual response power of the load cluster in the distribution area can be simplified into a deterministic constraint after being calculated using the expectation formula of the triangular fuzzy number. This constraint can characterize the uncertainty of the response of different load clusters. The actual response power of the converted load cluster can be expressed as the following formula.

[0111]

[0112] Among them, P′ IL (t) represents the response power of the load cluster that can be reduced at time t after uncertainty analysis, P′ SL (t) represents the response power of the shiftable load cluster at time t after uncertainty analysis, P′ TL (t) represents the response power of the transferable load cluster at time t after uncertainty analysis.

[0113]

[0114]

[0115]

[0116] For example, assume that the proportionality coefficient between the maximum error level and the rate of change of the compensation electricity price is k1 = 0.25 and k2 = -0.2. Table 3 shows the uncertainty parameters of the cluster response, and Table 4 shows the load cluster response power after taking uncertainty into account.

[0117] Table 3

[0118]

[0119]

[0120] Table 4

[0121]

[0122] Compared with other similar technologies, the advantages of this invention are as follows: By constructing a multi-dimensional time-series model of user response, it can effectively handle the uncertainty of load adjustment caused by differences in users' subjective intentions, improve the user participation rate in demand response, and reduce the response error rate. Furthermore, this invention uses a cluster-based user load hierarchical aggregation method to cluster and divide numerous response resources with small response capacity within a distribution area. This invention can effectively solve problems such as high cloud computing pressure caused by complex distribution area resources, and shorten computing time.

[0123] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0124] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0125] Figure 7 A schematic diagram of a control device for a power distribution network provided in an embodiment of the present invention is shown. The control device 300 includes a communication module 301 and a processing module 302.

[0126] The communication module 301 is used to obtain the time-of-use compensation price of the distribution network demand response and the response information declared by the power users themselves. The response information includes load type, response method, preferred time and response power.

[0127] Processing module 302 is used to determine the compensation cost of each power user based on time-of-use compensation price and response information; based on the compensation cost and response information of each power user, it determines the multi-dimensional time series model of each power user. The multi-dimensional time series model is used to characterize the time series characteristics of compensation cost and response power corresponding to each load type; based on the multi-dimensional time series model of each power user, it classifies and aggregates the demand response of each power user to obtain a cluster aggregation model to predict various types of demand response in the distribution network.

[0128] In one possible implementation, the processing module 302 is specifically used to determine the unit capacity compensation cost based on the time-of-use compensation price and a preset response mode coefficient, wherein the response mode coefficient is positively correlated with the flexibility of demand response; to determine the unit time compensation cost based on the time-of-use compensation price, preferred time, and expected response time; and to determine the compensation cost for each electricity user based on the unit capacity compensation cost and the unit time compensation cost.

[0129] In one possible implementation, the processing module 302 is specifically used to, for any power user, determine the change of compensation cost over time based on the power user's compensation cost; determine the change of response power over time for each load type based on the power user's load type, response method, preferred time, and response power; and determine the multi-dimensional time series model of the power user based on the change of compensation cost over time and the change of response power over time for each load type, wherein the multi-dimensional time series model of the power user represents the time series using a binary tree hierarchical structure, the response power uses the node function of the binary tree, and the compensation cost uses the inter-node function of the binary tree.

[0130] In one possible implementation, the processing module 302 is specifically used to classify and aggregate load types, compensation costs, and response power in a multi-dimensional time series model as feature variables to obtain sub-load clusters. Each sub-load cluster represents the sum of response power of multiple power users in a type of demand response. Based on each sub-load cluster, a cluster aggregation model is obtained. The cluster aggregation model includes load reduction clusters, load shifting clusters, and load transfer clusters, wherein each load cluster includes multiple sub-load clusters.

[0131] In one possible implementation, the processing module 302 is further configured to calculate the load response rate corresponding to each load type based on each sub-load cluster; calculate the uncertainty coefficient corresponding to each load type based on the load response rate corresponding to each load type; the uncertainty coefficient is used to characterize the magnitude of the response power deviation caused by the uncertainty of the demand response; and correct each sub-load cluster based on the uncertainty coefficient corresponding to each load type to obtain the corrected sub-load clusters.

[0132] Figure 8 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. For example... Figure 8 As shown, the electronic device 400 of this embodiment includes: a processor 401, a memory 402, and a computer program 403 stored in the memory 402 and executable on the processor 401. When the processor 401 executes the computer program 403, it implements the steps in the above-described method embodiments, for example... Figure 1 Steps 101 to 104 are shown. Alternatively, when the processor 401 executes the computer program 403, it implements the functions of each module / unit in the above-described device embodiments, for example... Figure 7 The functions of the communication module 301 and the processing module 302 shown are illustrated.

[0133] For example, the computer program 403 can be divided into one or more modules / units, which are stored in the memory 402 and executed by the processor 401 to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 403 in the electronic device 400. For example, the computer program 403 can be divided into... Figure 7 The communication module 301 and the processing module 302 are shown.

[0134] The processor 401 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0135] The memory 402 can be an internal storage unit of the electronic device 400, such as a hard disk or memory of the electronic device 400. The memory 402 can also be an external storage device of the electronic device 400, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 400. Furthermore, the memory 402 can include both internal and external storage units of the electronic device 400. The memory 402 is used to store the computer program and other programs and data required by the terminal. The memory 402 can also be used to temporarily store data that has been output or will be output.

[0136] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0137] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0138] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0139] In the embodiments provided by this invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0140] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0141] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0142] If the integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0143] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for predicting demand response of a power distribution network, characterized in that, The method comprises: obtaining time-of-use compensation price of demand response of a power distribution network and response information reported by power users, the response information including load type, response mode, preferred time and response power; determining compensation cost of each power user based on the time-of-use compensation price and the response information; determining a multi-dimensional time sequence model of each power user based on the compensation cost and the response information of each power user, including: for any power user, determining the change of the compensation cost with time based on the compensation cost of the power user; determining the change of each load type corresponding response power with time based on the load type, response mode, preferred time and response power of the power user; determining the multi-dimensional time sequence model of the power user based on the change of the compensation cost with time and the change of each load type corresponding response power with time, wherein the multi-dimensional time sequence model of the power user represents time sequence in the form of link level structure of a binary tree, represents response power in the form of node function of a binary tree, and represents compensation cost in the form of function between nodes of a binary tree; the multi-dimensional time sequence model is used to represent time sequence characteristics of compensation cost and time sequence characteristics of response power corresponding to each load type; classifying and aggregating demand response of each power user based on the multi-dimensional time sequence model of each power user to obtain a cluster aggregation model to predict each type of demand response in the power distribution network.

2. The method of claim 1, wherein, The method further comprises: determining unit capacity compensation cost of each power user based on the time-of-use compensation price and the response information, including: determining the unit capacity compensation cost based on the time-of-use compensation price and a preset response mode coefficient, the response mode coefficient being positively correlated with flexibility of demand response; determining unit time compensation cost based on the time-of-use compensation price, preferred time and expected response time; 3. The method of claim 1, wherein, determining the compensation cost of each power user based on the unit capacity compensation cost and the unit time compensation cost. The method further comprises: performing classification and aggregation based on load type, compensation cost and response power in the multi-dimensional time sequence model to obtain each sub-load cluster, each sub-load cluster representing the sum of response power of multiple power users in a type of demand response; 4. The method of claim 3, wherein, aggregating the sub-load clusters to obtain the cluster aggregation model, the cluster aggregation model including a reducible load cluster, a shiftable load cluster and a transferable load cluster, wherein each load cluster includes multiple sub-load clusters. The method further comprises: calculating load response rate corresponding to each load type based on each sub-load cluster; calculating an uncertainty coefficient corresponding to each load type based on the load response rate corresponding to each load type, the uncertainty coefficient being used to represent the size of response power deviation caused by uncertainty of demand response; 5. A control device of a power distribution network, characterized by, modifying each sub-load cluster based on the uncertainty coefficient corresponding to each load type to obtain modified each sub-load cluster. The method comprises: The communication module is configured to acquire a time-of-use compensation price of a demand response of a power distribution network and response information reported by a power user, the response information including a load type, a response mode, a preferred time, and a response power; The processing module is configured to determine a compensation cost of each power user based on the time-of-use compensation price and the response information. Based on the compensation cost and the response information of each power user, a multi-dimensional time sequence model of each power user is determined, the multi-dimensional time sequence model being used to represent time sequence characteristics of the compensation cost and time sequence characteristics of the response power corresponding to each load type; and based on the multi-dimensional time sequence model of each power user, the demand response of each power user is classified and aggregated to obtain a cluster aggregation model, so as to predict various demand responses in the power distribution network. The processing module is specifically configured to, for any power user, determine a change of the compensation cost with time based on the compensation cost of the power user; and determine a change of the response power with time corresponding to each load type based on the load type, the response mode, the preferred time, and the response power of the power user. Based on the change of the compensation cost with time and the change of the response power with time corresponding to each load type, the multi-dimensional time sequence model of the power user is determined, wherein the multi-dimensional time sequence model of the power user represents a time sequence in a link level structure of a binary tree, represents the response power in a node function of the binary tree, and represents the compensation cost in a function between nodes of the binary tree.

6. The control device of the power distribution network according to claim 5, wherein The processing module is specifically configured to determine a unit capacity compensation cost based on the time-of-use compensation price and a preset response mode coefficient, the response mode coefficient being positively correlated with flexibility of the demand response; determine a unit time compensation cost based on the time-of-use compensation price, the preferred time, and an expected response time; and determine the compensation cost of each power user based on the unit capacity compensation cost and the unit time compensation cost.

7. An electronic device, comprising: The electronic device includes a memory and a processor, the memory storing a computer program, and the processor being configured to invoke and run the computer program stored in the memory to execute the method according to any one of claims 1 to 4.

8. A computer-readable storage medium storing a computer program, the computer-readable storage medium comprising: The computer program is executed by the processor to implement the steps of the method according to any one of claims 1 to 4.

Images