CN118691116A - A county load aggregation quantification method and system - Google Patents

Abstract

The present invention relates to the field of power grid regulation technology, and in particular to a method and system for quantifying county load aggregation, including clustering analysis of county historical load curves to obtain load power consumption patterns; based on the load power consumption patterns of the regulation day and the load baseline of the regulation day, the adjustable amount of the county load is obtained; based on the obtained load power consumption uncertainty representation and the adjustable amount of the county load, a single load adjustable potential sequence is obtained through discretization to obtain the probability distribution of the total load adjustable potential of the feeder line in the substation area, and based on the probability distribution of the total load adjustable potential of the feeder line in the substation area, it is iteratively updated using a load aggregation algorithm to obtain the adjustable potential distribution of each load aggregation cluster. The method proposed in the present invention realizes the accurate dynamic evaluation of the resource regulation potential on the user side, and more effectively optimizes resource allocation and improves energy utilization efficiency through the complementary characteristics of the county characteristic load and distributed power sources in the time and space dimensions.

Description

A county load aggregation quantification method and system

Technical Field

The present invention relates to the technical field of power grid regulation, and in particular to a county load aggregation quantification method and system.

Background Art

The green transformation of energy in rural areas of counties is a key link in the construction of a modern energy system. Its characteristic loads, such as agricultural irrigation, electric tobacco roasting, electric tea roasting and heat storage electric boilers, have the remarkable characteristics of decentralized access, wide distribution and strong seasonal fluctuations. These loads not only have a high degree of electricity flexibility, but also have great adjustable potential. At the same time, the wide access of distributed power sources within the county has further enriched the energy structure.

In the strategy of improving system flexibility resources, load aggregation and quantification technology has been widely used in cluster research such as electric vehicles, temperature-controlled loads and micro-energy networks. For example, the daily load curve is classified and aggregated through deep learning algorithms and frequency-domain discrete Fourier transform, and the aggregate characteristics of multiple types of load users such as agricultural users are extracted; other scholars have proposed a clustering method based on dimensionality reduction of characteristic indicators, and used the Delphi method based on clustering validity correction to configure the weights of each indicator to achieve accurate aggregation and quantification of the daily load curve. At the same time, other studies have combined hierarchical analysis and regression analysis methods to achieve industry-wide load aggregation in county-level power departments.

Although the country has carried out a lot of practice in user load aggregation and quantification, and verified the significant effect of the synergy between user load and distributed power sources, facing the demand for building a strong smart grid in the future, the aggregation and quantification of county-level characteristic loads still faces the following problems that need to be solved:

(1) The electricity consumption behavior of county users is affected by multiple factors, such as price, usage habits, parameters of household appliances, and various environmental factors. It is highly uncertain and difficult to achieve unified evaluation and aggregated regulation while ensuring the usage experience of most users.

(2) At present, the penetration rate of smart terminals among county users is low, and it is difficult to obtain information on adjustable electrical appliances with characteristic loads in counties. Traditional electrical appliance modeling methods have large errors in potential assessment, which makes it difficult to effectively match the adjustable characteristics of characteristic loads with distributed power sources.

(3) Existing research focuses on the energy demand characteristics of a single industry scenario of new agriculture, lacking research on the differentiated energy demand of county-level characteristic loads such as modern agriculture and characteristic industries in typical regions, which limits the comprehensiveness and depth of the aggregate quantification of county-level characteristic loads;

(4) Current research focuses on the multidimensional characteristics of new production and living loads in counties, and concentrates on classification and aggregation methods and models based on mathematical and physical dimensions. There is a lack of in-depth analysis of the timing characteristics of county-level characteristic loads and the flexible response characteristics of production and living energy facilities. At the same time, there is also a lack of decentralized regulation and complementary aggregation of specific county-level characteristic loads and distributed power sources in the temporal and spatial dimensions.

Therefore, there is an urgent need to provide a county load aggregation quantification method to achieve efficient and sustainable development of energy green transformation in county rural areas.

Summary of the invention

The purpose of the present invention is to provide a county load aggregation quantification method and system to achieve accurate dynamic evaluation of user-side resource regulation potential, and to more effectively optimize resource allocation and improve energy utilization efficiency through the complementary characteristics of county characteristic loads and distributed power sources in time and space dimensions.

In a first aspect, the present invention provides a method for quantifying county load aggregation, the method comprising the following steps:

Obtain the historical load curve of the county, and obtain the load power consumption pattern by clustering the historical load curve of the county;

Determine the load power consumption mode on the control day, and obtain the adjustable load amount in the county based on the load power consumption mode on the control day and the load baseline on the control day;

Calculate the uncertainty characterization of load power consumption, and obtain a single load adjustable potential sequence through discretization based on the uncertainty characterization of load power consumption and the adjustable amount of county load; wherein the uncertainty characterization of load power consumption includes the probability of a single load participating in the regulation task in a certain period of time and the user adjustable load rate in the regulation period of the regulation day;

According to the adjustable potential sequence of a single load of the low-voltage feeder during the regulation period of the regulation day, the probability distribution of the total load adjustable potential of the feeder in the substation area is obtained;

The adjustable potential probability distribution of the total load in the substation is integrated with the pre-collected real-time data of the feeder load in the substation into adjustable potential load data, and the adjustable potential load data is iteratively updated using a pre-selected load aggregation algorithm to obtain the adjustable potential distribution of each load aggregation cluster.

In a further implementation scheme, the step of determining the load power consumption mode of the control day and obtaining the adjustable amount of the county load based on the load power consumption mode of the control day and the load baseline of the control day includes:

Determine the load baseline for the regulation day according to the daily load curves of several similar days before the regulation day, and calculate the load mode membership for the regulation day according to the load baseline for the regulation day and the typical load curves of different load power consumption modes;

The load power consumption mode corresponding to the maximum load mode membership degree on the control day is selected as the power consumption mode of the control day, and the load adjustable potential of the control day regulation period is calculated according to the load power consumption mode of the control day;

According to the load adjustable potential during the regulation period of the regulation day and the load baseline of the regulation day, the adjustable amount of the county load is calculated; the adjustable amount of the county load includes the most likely adjustable amount of the county load and the maximum adjustable amount of the county load.

In a further embodiment, the mathematical expression of the load adjustable potential is:

Where ΔP _m,j (t) represents the load adjustable potential in the regulation period t of the regulation day; It represents the load baseline in the regulation period t of the regulation day; It represents the load curve with the longest Euclidean distance from the load baseline of the regulation day under the same load power consumption mode as the load baseline of the regulation day; m represents the load power consumption mode to which the load baseline of the regulation day belongs; j represents the load power consumption mode after the regulation event occurs; z _φ represents the load curve under the low load power consumption mode within the regulation period t.

In a further embodiment, the mathematical expression of the most likely adjustment amount of the county load is:

In the formula, ΔP _α (t) represents the most likely adjustment amount of county load; It represents the load baseline in the regulation period t of the regulation day; It indicates the expected minimum load under the load power consumption mode during the regulation period of the regulation day;

The mathematical expression of the maximum adjustment amount of the county load is:

In the formula, ΔP _β (t) represents the maximum regulation of county load; represents the load baseline within the regulation period t of the regulation day; ^h2 represents the typical load curve under the lowest load power consumption mode among all load power consumption modes.

In a further embodiment, the step of calculating the load power consumption uncertainty characterization and obtaining a single load adjustable potential sequence by discretization based on the load power consumption uncertainty characterization and the county load adjustable amount includes:

According to the most likely load regulation amount in the county and the actual load regulation of users on similar days in the historical regulation behavior tags, the actual load regulation rate during the regulation period of similar days is obtained;

According to the actual load regulation rate of the regulation period on similar days and the number of regulation tasks in the same regulation period on similar days, the user adjustable load rate of the regulation period on the regulation day is calculated;

The probability of a single load participating in the regulation task in a certain period of time is calculated according to the response sensitivity coefficient, and the adjustable potential vector is constructed according to the probability of a single load participating in the regulation task in a certain period of time, the user adjustable load rate in the regulation period of the regulation day, and the adjustable amount of the county load.

According to the sequence operation theory, the adjustable potential vector is discretized into a single load adjustable potential sequence.

In a further embodiment, the mathematical expression of the user adjustable load rate during the regulation period of the regulation day is:

in,

Where, χ(t) represents the user's adjustable load rate during the regulation period of the regulation day; _χi represents the actual load regulation rate during the regulation period on similar days; _γi represents the identifier of the user's successful participation in the demand regulation task; n represents the number of regulation tasks during the same period on similar days; _ΔPi represents the actual load regulation of users on similar days in the historical regulation behavior label; ΔP _α (t) represents the most likely adjustment amount of the user's county load during the regulation period;

The mathematical expression of the probability of a single load participating in the regulation task in a certain period of time is:

Where p(t) represents the probability of a single load participating in the regulation task in a certain period of time; ω represents the sensitivity of the load to participate in the conditional task; τ represents the response sensitivity coefficient; K represents the normalized mean of the historical regulation evaluation score after the user makes effective regulation, where 1≤K≤5.

In a further embodiment, the single load adjustable potential sequence should meet the following conditions:

Where ΔP _α (t) represents the most likely adjustment amount of the county load during the adjustment period; χ(t) represents the adjustable load rate of the user during the adjustment period of the regulation day; p(t) represents the probability of a single load participating in the adjustment task during a certain period; A _q represents the adjustable potential sequence of a single load; and _Na represents the sequence length, that is, the theoretical adjustable potential of the load.

In a further embodiment, the step of obtaining the probability distribution of the total load adjustable potential of the feeder in the substation area according to the single load adjustable potential sequence of the low-voltage feeder in the regulation period of the regulation day includes:

According to the adjustable potential sequence of a single load of the low-voltage feeder during the regulation period of the regulation day, the probability distribution of the total adjustable potential of the low-voltage feeder load is constructed;

The probability distribution of the total load adjustable potential of each low-voltage feeder in the substation during the regulation period is rolled and calculated to obtain the probability distribution of the total load adjustable potential of the feeder in the substation.

In a further embodiment, the step of iteratively updating the adjustable potential load data using a pre-selected load aggregation algorithm to obtain the adjustable potential distribution of each load aggregation cluster includes:

The K-means algorithm is selected as the load aggregation algorithm, and the load aggregation convergence target and load aggregation scale are determined according to the adjustable potential load data;

Several samples are randomly selected from the adjustable potential load data as the initial aggregation centers, and the load aggregation convergence target and load aggregation scale are determined according to the adjustable potential load data. The load aggregation algorithm is used to cluster the adjustable potential load data. After several iterations, the adjustable potential distribution of each load aggregation cluster is obtained.

In a second aspect, the present invention provides a county load aggregation quantification system, the system comprising:

The load pattern analysis module is used to obtain the historical load curve of the county and obtain the load power consumption pattern by clustering the historical load curve of the county;

The load regulation amount acquisition module is used to determine the load power consumption mode of the regulation day, and obtain the county load adjustable amount based on the load power consumption mode of the regulation day and the load baseline of the regulation day;

A single load potential analysis module is used to calculate the uncertainty characterization of load power consumption, and based on the uncertainty characterization of load power consumption and the adjustable amount of county load, obtain a single load adjustable potential sequence through discretization; wherein, the uncertainty characterization of load power consumption includes the probability of a single load participating in the regulation task in a certain period of time and the user adjustable load rate in the regulation period of the regulation day;

The multi-load potential analysis module is used to obtain the probability distribution of the total load adjustable potential of the feeder in the substation area according to the single load adjustable potential sequence of the low-voltage feeder during the regulation period of the regulation day;

The load aggregation quantification module is used to integrate the probability distribution of the adjustable potential of the total load in the substation and the real-time data of the feeder load in the substation collected in advance into adjustable potential load data, and use the pre-selected load aggregation algorithm to iteratively update the adjustable potential load data to obtain the adjustable potential distribution of each load aggregation cluster.

In a further embodiment, the load adjustment amount acquisition module is specifically used to:

Determine the load baseline for the regulation day according to the daily load curves of several similar days before the regulation day, and calculate the load mode membership for the regulation day according to the load baseline for the regulation day and the typical load curves of different load power consumption modes;

The load power consumption mode corresponding to the maximum load mode membership degree on the control day is selected as the power consumption mode of the control day, and the load adjustable potential of the control day regulation period is calculated according to the load power consumption mode of the control day;

According to the load adjustable potential during the regulation period of the regulation day and the load baseline of the regulation day, the adjustable amount of the county load is calculated; the adjustable amount of the county load includes the most likely adjustable amount of the county load and the maximum adjustable amount of the county load.

In a further embodiment, the mathematical expression of the load adjustable potential is:

Where ΔP _m,j (t) represents the load adjustable potential in the regulation period t of the regulation day; It represents the load baseline in the regulation period t of the regulation day; It represents the load curve with the longest Euclidean distance from the load baseline of the regulation day under the same load power consumption mode as the load baseline of the regulation day; m represents the load power consumption mode to which the load baseline of the regulation day belongs; j represents the load power consumption mode after the regulation event occurs; z _φ represents the load curve under the low load power consumption mode within the regulation period t.

In a further embodiment, the mathematical expression of the most likely adjustment amount of the county load is:

In the formula, ΔP _α (t) represents the most likely adjustment amount of county load; It represents the load baseline in the regulation period t of the regulation day; It indicates the expected minimum load under the load power consumption mode during the regulation period of the regulation day;

The mathematical expression of the maximum adjustment amount of the county load is:

In the formula, ΔP _β (t) represents the maximum regulation of county load; represents the load baseline within the regulation period t of the regulation day; ^h2 represents the typical load curve under the lowest load power consumption mode among all load power consumption modes.

In a further embodiment, the single load potential analysis module is specifically used for:

According to the most likely load regulation amount in the county and the actual load regulation of users on similar days in the historical regulation behavior tags, the actual load regulation rate during the regulation period of similar days is obtained;

According to the actual load regulation rate of the regulation period on similar days and the number of regulation tasks in the same regulation period on similar days, the user adjustable load rate of the regulation period on the regulation day is calculated;

The probability of a single load participating in the regulation task in a certain period of time is calculated according to the response sensitivity coefficient, and the adjustable potential vector is constructed according to the probability of a single load participating in the regulation task in a certain period of time, the user adjustable load rate in the regulation period of the regulation day, and the adjustable amount of the county load.

According to the sequence operation theory, the adjustable potential vector is discretized into a single load adjustable potential sequence.

In a further embodiment, the mathematical expression of the user adjustable load rate during the regulation period of the regulation day is:

in,

Where, χ(t) represents the user's adjustable load rate during the regulation period of the regulation day; _χi represents the actual load regulation rate during the regulation period on similar days; _γi represents the identifier of the user's successful participation in the demand regulation task; n represents the number of regulation tasks during the same period on similar days; _ΔPi represents the actual load regulation of users on similar days in the historical regulation behavior label; ΔP _α (t) represents the most likely adjustment amount of the user's county load during the regulation period;

The mathematical expression of the probability of a single load participating in the regulation task in a certain period of time is:

Where p(t) represents the probability of a single load participating in the regulation task in a certain period of time; ω represents the sensitivity of the load to participate in the conditional task; τ represents the response sensitivity coefficient; K represents the normalized mean of the historical regulation evaluation score after the user makes effective regulation, where 1≤K≤5.

In a further embodiment, the single load adjustable potential sequence should meet the following conditions:

Where ΔP _α (t) represents the most likely adjustment amount of the county load during the adjustment period; χ(t) represents the adjustable load rate of the user during the adjustment period of the regulation day; p(t) represents the probability of a single load participating in the adjustment task during a certain period; A _q represents the adjustable potential sequence of a single load; and _Na represents the sequence length, that is, the theoretical adjustable potential of the load.

In a further embodiment, the multi-load potential analysis module is specifically used for:

According to the adjustable potential sequence of a single load of the low-voltage feeder during the regulation period of the regulation day, the probability distribution of the total adjustable potential of the low-voltage feeder load is constructed;

The probability distribution of the total load adjustable potential of each low-voltage feeder in the substation during the regulation period is rolled and calculated to obtain the probability distribution of the total load adjustable potential of the feeder in the substation.

In a further embodiment, the iterative updating of the adjustable potential load data using a pre-selected load aggregation algorithm to obtain the adjustable potential distribution of each load aggregation cluster specifically includes:

The K-means algorithm is selected as the load aggregation algorithm, and the load aggregation convergence target and load aggregation scale are determined according to the adjustable potential load data;

Several samples are randomly selected from the adjustable potential load data as the initial aggregation centers, and the load aggregation convergence target and load aggregation scale are determined according to the adjustable potential load data. The load aggregation algorithm is used to cluster the adjustable potential load data. After several iterations, the adjustable potential distribution of each load aggregation cluster is obtained.

The present invention provides a method and system for quantifying county load aggregation. The method obtains a load power consumption pattern by clustering analysis of county historical load curves; determines the load power consumption pattern of a regulation day, and obtains the county load adjustable amount based on the load power consumption pattern of the regulation day and the load baseline of the regulation day; obtains a single load adjustable potential sequence by discretization based on the obtained load power consumption uncertainty representation and the county load adjustable amount; obtains the total load adjustable potential probability distribution of the substation feeder according to the single load adjustable potential sequence of the low-voltage feeder within the regulation period of the regulation day; integrates the total load adjustable potential probability distribution of the substation with the pre-collected real-time data of the substation feeder load into adjustable potential load data, and uses a load aggregation algorithm to iteratively update the adjustable potential load data to obtain the adjustable potential distribution of each load aggregation cluster. Compared with the existing technology, this method proposes a characterization of county-specific load regulation uncertainty including the regulation load rate and regulation probability, and based on the application of sequence operation theory in load adjustable potential analysis, combines the K-means aggregation algorithm to establish a county-specific load aggregation quantification method based on adjustable potential evaluation. Through accurate load adjustable potential evaluation and optimized load aggregation strategy, accurate dynamic analysis of user-side resource regulation potential can be achieved, and the spatiotemporal complementary characteristics of county-specific loads and distributed power sources can be mastered, thereby realizing real-time monitoring and regulation of the power grid operation status and improving the reliability and stability of the power grid.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a method for quantifying county load aggregation provided by an embodiment of the present invention;

FIG2 is a schematic diagram of a load adjustable potential curve provided by an embodiment of the present invention;

FIG3 is a schematic diagram of a single load adjustable potential sequence provided by an embodiment of the present invention;

FIG4 is a schematic diagram of a distribution network structure provided by an embodiment of the present invention;

FIG5 is a schematic diagram of a K-means algorithm flow chart provided by an embodiment of the present invention;

FIG6 is a schematic diagram of an adjustable capacity curve of load A provided in an embodiment of the present invention;

7 is a schematic diagram of an adjustable capacity curve of load B provided in an embodiment of the present invention;

FIG8 is a schematic diagram of the probability distribution of the adjustable potential of the feeder w at the time point 18:30 provided by an embodiment of the present invention;

9 is a schematic diagram of the fitting result of the probability distribution of adjustable potential of the station area 1 at the time point 18:30 provided by an embodiment of the present invention;

10 is a schematic diagram of the fitting result of the probability distribution of the adjustable potential of the station area I at the time point 18:30 provided in an embodiment of the present invention;

11 is a schematic diagram of the fitting result of the probability distribution of adjustable potential of the station area I at the time point 18:30 provided by an embodiment of the present invention;

FIG12 is a block diagram of a county load aggregation and quantification system provided in an embodiment of the present invention.

DETAILED DESCRIPTION

The following specifically illustrates the implementation mode of the present invention in conjunction with the accompanying drawings. The embodiments are provided for illustrative purposes only and are not to be construed as limitations of the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute limitations on the scope of patent protection of the present invention, because many changes may be made to the present invention without departing from the spirit and scope of the present invention.

Referring to FIG1 , an embodiment of the present invention provides a method for quantifying county load aggregation. As shown in FIG1 , the method includes the following steps:

S1. Obtain the historical load curve of the county, and obtain the load power consumption pattern by performing cluster analysis on the historical load curve of the county.

The historical load curve of the characteristic load in the county collected by the advanced measurement system in this embodiment can reflect the power consumption behavior of the load, which is the result of the explicit performance of the decision made by the user on the influencing factors. It should be noted that the influencing factors of the characteristic load power consumption behavior and the sensitivity to these factors are implicit in the historical load curve. Therefore, this embodiment can identify the potential power consumption mode of the load by performing aggregation analysis on the historical load curve of the load. The load power consumption mode represents a cluster formed by a group of identical or similar historical load curves. The specific representation is as follows:

In the formula, c _k represents the cluster of electricity consumption mode k; represents the jth load curve of power consumption mode k; M represents the sampling times of the daily load curve.

The historical load curve set can be expressed as a set of multiple power consumption modes of the load. The specific representation is as follows:

D _N = c ₁ + c ₂ + … + c _K

Where D _N represents the daily load curve for N consecutive days, usually one year, one season or one month; K represents the number of load power consumption modes.

When the power system sends a demand response task or regulation signal to a user, the user will adjust his or her own load under constraints according to the task requirements to achieve power mode switching and load regulation. Therefore, the power mode can be identified by analyzing the load's historical load curve, and the feasibility analysis of power mode switching can be combined to evaluate the overall adjustable potential of the characteristic load.

S2. Determine the load power consumption pattern on the control day, and obtain the adjustable load amount in the county based on the load power consumption pattern on the control day and the load baseline on the control day.

In this embodiment, the step of determining the load power consumption mode of the control day and obtaining the adjustable amount of the county load based on the load power consumption mode of the control day and the load baseline of the control day includes:

Determine the load baseline for the regulation day according to the daily load curves of several similar days before the regulation day, and calculate the load mode membership for the regulation day according to the load baseline for the regulation day and the typical load curves of different load power consumption modes;

The load power consumption mode corresponding to the maximum load mode membership degree on the control day is selected as the power consumption mode of the control day, and the load adjustable potential of the control day regulation period is calculated according to the load power consumption mode of the control day;

According to the load adjustable potential during the regulation period of the regulation day and the load baseline of the regulation day, the adjustable amount of the county load is calculated; the adjustable amount of the county load includes the most likely adjustable amount of the county load and the maximum adjustable amount of the county load.

Specifically, in this embodiment, the load baseline of the control day is used to determine the power consumption mode of the load on that day. The load baseline of the control day is calculated as follows:

In the formula, represents the load baseline of the control day; _xi represents the daily load curve of the i-th similar day before the control day. When the control day is a normal working day, n=5; when the control day is a rest day, n=3.

The degree of membership between the control day and each load power consumption mode is shown in the following formula:

In the formula, To control the daily load baseline Membership degree to load power consumption mode k ₂ ; is the typical load curve of load power consumption mode k ₁ (aggregate cluster aggregation center); is the typical load curve of load power consumption mode k ₁ (aggregate cluster aggregation center).

The electricity consumption mode with the highest degree of membership on the control day is as follows:

As shown in FIG2 , this embodiment sets the load baseline in the regulation period t=[t _start ,t _end ] of the regulation day as The load adjustable potential ΔP _m,j (t) during this period is defined as:

Where ΔP _m,j (t) represents the load adjustable potential in the regulation period t of the regulation day; represents the load baseline in the regulation period t of the regulation day; z _φ represents the load curve in the low-load power consumption mode in the regulation period t (daily load curve 1); It represents the load curve (daily load curve 2) with the longest Euclidean distance from the control day load baseline under the same load power consumption mode as the control day load baseline; m represents the load power consumption mode to which the control day load baseline belongs; j represents the load power consumption mode after the control event occurs.

In this embodiment, the load adjustment amount under different power consumption modes in the regulation and control period t can be obtained according to the above load adjustment potential calculation formula, and the following two indicators are defined to describe the load adjustment potential: the most likely adjustment amount ΔP _α (t) of the county load, and the maximum adjustment amount ΔP _β (t) of the county load, wherein the most likely adjustment amount ΔP _α (t) does not require the load to switch the power consumption mode, and is the most likely adjustment method for the load. The specific calculation method is the difference between the baseline load in the regulation period and the lowest load curve in the working mode to which it belongs; the maximum adjustment amount ΔP _β (t) requires the load to change the power consumption mode to the maximum extent, and is the adjustment method with the highest load participation. The specific calculation method is the difference between the baseline load in the regulation period and the typical load curve under the working mode with the lowest load. In this embodiment, the mathematical expression of the most likely adjustment amount of the county load is:

In the formula, ΔP _α (t) represents the most likely adjustment amount of county load; It represents the load baseline in the regulation period t of the regulation day; It indicates the minimum load expected under the load power consumption mode during the regulation period of the regulation day.

The mathematical expression of the maximum adjustment amount of the county load is:

In the formula, ΔP _β (t) represents the maximum regulation of county load; represents the load baseline within the regulation period t of the regulation day; ^h2 represents the typical load curve under the lowest load power consumption mode among all load power consumption modes.

S3. Calculate the uncertainty characterization of load power consumption, and based on the uncertainty characterization of load power consumption and the adjustable amount of county load, obtain a single load adjustable potential sequence through discretization; wherein, the uncertainty characterization of load power consumption includes the probability of a single load participating in the regulation task in a certain period of time and the user adjustable load rate in the regulation period of the regulation day.

At present, sequence operation theory is a mathematical tool for solving complex discrete probability problems. It originated from sequence convolution in the field of digital signals. After rigorous mathematical abstraction and improvement, it has been widely used in the field of power system planning. With its advantages of simple calculation and clear concepts, sequence operation theory has become a powerful tool for solving complex problems such as random production simulation and uncertain electricity price analysis in power systems. For owners of county-level characteristic loads, the uncertainty of load power consumption is reflected in two aspects: load regulation rate and regulation willingness. The user's participation in the regulation task can be detected through an advanced measurement system, and the user's historical regulation situation and related label information are used to characterize the uncertain behavior. In this embodiment, the calculation of the load power consumption uncertainty characterization, and based on the load power consumption uncertainty characterization and the county load adjustable amount, the steps of obtaining a single load adjustable potential sequence through discretization include:

According to the most likely load regulation amount in the county and the actual load regulation of users on similar days in the historical regulation behavior tags, the actual load regulation rate during the regulation period of similar days is obtained;

According to the actual load regulation rate of the regulation period on similar days and the number of regulation tasks in the same regulation period on similar days, the user adjustable load rate of the regulation period on the regulation day is calculated;

The probability of a single load participating in the regulation task in a certain period of time is calculated according to the response sensitivity coefficient, and the adjustable potential vector is constructed according to the probability of a single load participating in the regulation task in a certain period of time, the user adjustable load rate in the regulation period of the regulation day, and the adjustable amount of the county load.

According to the sequence operation theory, the adjustable potential vector is discretized into a single load adjustable potential sequence.

Specifically, the actual load regulation rate of the regulation period t in the similar day i defined in this embodiment can be expressed as

Where χ _i represents the actual load regulation rate during the regulation period on similar days; ΔP _i represents the actual load regulation of users on similar days in the historical regulation behavior label; ΔP _α (t) represents the most likely load regulation amount of the user in the county during the regulation period.

The user adjustable load rate χ(t) within the regulation period t = [t _start , t _end ] within the regulation day can be expressed as:

Where χ(t) represents the user adjustable load rate during the regulation period of the regulation day; γ _i represents the user's successful participation in the demand regulation task, that is, whether the user successfully participates in the demand regulation task; n represents the number of regulation tasks in the same period on similar days, which is generally 5.

In this embodiment, the probability of a single load participating in the regulation task in a certain period of time is calculated by combining the historical regulation behavior labels of the load users as shown in the following formula:

Where p(t) represents the probability of a single load participating in the regulation task in a certain period of time; ω represents the sensitivity of the load to participate in the conditional task; τ represents the response sensitivity coefficient, which is generally taken as τ=1, and technical personnel in this field can adjust it according to the historical regulation behavior; K represents the normalized mean of the historical regulation evaluation score after the user makes effective regulation, where 1≤K≤5.

Finally, this embodiment integrates the load regulation amount and the four indicators of uncertainty into an adjustable potential vector to comprehensively represent the load adjustable potential:

R _g =[ΔP _α (t),ΔP _β (t),χ(t),p(t)]

According to the sequence operation theory, the adjustable potential of a single user q can be discretized into a simple probability sequence A _q as shown in Figure 3. The adjustable potential sequence A _q of a single load should meet the following conditions:

Where ΔP _α (t) represents the most likely adjustment amount of the county load during the adjustment period; χ(t) represents the adjustable load rate of the user during the adjustment period of the regulation day; p(t) represents the probability of a single load participating in the adjustment task during a certain period; A _q represents the adjustable potential sequence of a single load; and _Na represents the sequence length, that is, the theoretical adjustable potential of the load.

S4. According to the adjustable potential sequence of a single load of the low-voltage feeder during the regulation period of the regulation day, the probability distribution of the total adjustable potential load of the feeder in the substation is obtained.

In this embodiment, the step of obtaining the probability distribution of the total load adjustable potential of the feeder in the substation area according to the single load adjustable potential sequence of the low-voltage feeder in the regulation period of the regulation day includes:

According to the adjustable potential sequence of a single load of the low-voltage feeder during the regulation period of the regulation day, the probability distribution of the total adjustable potential of the low-voltage feeder load is constructed;

The probability distribution of the total load adjustable potential of each low-voltage feeder in the substation during the regulation period is rolled and calculated to obtain the probability distribution of the total load adjustable potential of the feeder in the substation.

Normally, the adjustable potential between each load is independent of each other. By using the sequence operation theory and the adjustable potential sequences of different loads and adopting the derivative operation rules of probability sequences, the final probability sequence describing the adjustable potential of multiple loads is obtained through reasonable operation. In this embodiment, the distribution network structure is shown in FIG4. According to the grid structure, the total adjustable potential of regional loads is the superposition of the adjustable potential of loads in the region, that is, the volume and result of the adjustable potential sequence of loads in the region. In this embodiment, the potential sequences of q users of low-voltage feeder w in the regulation day t period are respectively Then the total load adjustable potential probability distribution of the feeder can be expressed as:

The probability distribution of the total load adjustable potential of the substation l in period t can be expressed as the volume sum of the adjustment capacity sequences of all feeders under the substation:

Where ⊕ represents the roll and operation symbol; _Fw (t) represents the probability distribution of the total load adjustable potential of low-voltage feeder w; _Hl (t) represents the probability distribution of the total load adjustable potential of feeders in the substation; W represents the number of feeders in the substation.

S5. Integrate the adjustable potential probability distribution of the total load in the substation area with the pre-collected real-time data of the feeder load in the substation area into adjustable potential load data, and use the pre-selected load aggregation algorithm to iteratively update the adjustable potential load data to obtain the adjustable potential distribution of each load aggregation cluster.

In this embodiment, the step of iteratively updating the adjustable potential load data using a pre-selected load aggregation algorithm to obtain the adjustable potential distribution of each load aggregation cluster includes:

The K-means algorithm is selected as the load aggregation algorithm, and the load aggregation convergence target and load aggregation scale are determined according to the adjustable potential load data;

Several samples are randomly selected from the adjustable potential load data as the initial aggregation centers, and the load aggregation convergence target and load aggregation scale are determined according to the adjustable potential load data. The load aggregation algorithm is used to cluster the adjustable potential load data. After several iterations, the adjustable potential distribution of each load aggregation cluster is obtained.

Specifically, in this field, improving traditional clustering methods according to application scenarios is the main research direction at present. The main processes include:

(1) Load data preprocessing, removing outliers and patching missing values;

(2) Normalize load data to improve computing efficiency;

(3) Select appropriate aggregation algorithm according to the application scenario;

(4) Select effectiveness indicators and evaluate the aggregation results.

The load aggregation of the power system can be realized by the K-means algorithm. The K-means algorithm is an aggregation method based on partitioning. As shown in Figure 5, the K-means algorithm first randomly selects K samples as the initial aggregation centers, and then organizes the surrounding samples into K clusters based on these centers, and calculates the centroid of each cluster (that is, the average value of all points in the cluster); then, it iteratively updates the aggregation center, and uses the new aggregation center as the reference point for the next round until the aggregation center of two adjacent iterations does not change. At this time, it indicates that the algorithm converges and the aggregation effect reaches the optimal. In the process of iterative updating, the sample points are classified into the cluster where the aggregation center closest to it is located, the objective function gradually converges, and the aggregation effect is getting better and better. In the process of load aggregation, by combining the probability distribution of the adjustable potential of the total load in the substation with the load data, not only can the effectiveness of the aggregation results be improved, but also it can ensure that the aggregation results can fully reflect the adjustable potential of different loads, and provide a more scientific and reasonable decision-making basis for subsequent load management and scheduling. The specific steps are as follows:

1) Combine the probability distribution of the total load adjustable potential of the substation area with the load data to form a complete data set, and standardize the integrated data set to eliminate the influence of different dimensions on the aggregation results. In this embodiment, it is assumed that the data sample set size is N, the number of aggregations is K and the convergence threshold, and the iteration number flag I=1. K samples are randomly selected from the normalized load data set as the initial aggregation center. During the iteration process, the probability distribution of the total load adjustable potential of the substation area can be used as a reference to ensure that the aggregation results can fully reflect the adjustable potential of different loads. For example, loads with similar adjustable potential can be classified into the same cluster;

2) Calculate the distance between each sample object and the aggregation center If satisfied:

Then the sample _xi is considered to belong to class _cj .

3) Calculate the aggregation criterion function and use the aggregation criterion function (such as the sum of squares of intra-cluster distances) to evaluate the aggregation effect;

4) If the value of the aggregation criterion function is less than the preset convergence threshold, or the aggregation center does not change much in two consecutive iterations, the algorithm is considered to have converged;

5) Output the sample set of K load aggregation clusters and their corresponding aggregation centers, as well as the adjustable potential distribution of each cluster.

The advantage of the K-means algorithm lies in the rapid aggregation processing of large data sets. Its algorithm complexity is O(NKI). The load aggregation algorithm adopted in this embodiment not only considers the load data itself in the load aggregation process, but also considers the adjustable potential of the load. In combination with the adjustable potential, the load characteristics of each aggregation cluster are analyzed to output the adjustable potential distribution of each aggregation cluster, thereby improving the accuracy and practicality of the aggregation, better optimizing the management of county-level characteristic loads, and optimizing the operation of the power grid and the load scheduling strategy.

In order to verify the county load aggregation quantification method proposed in this embodiment, the following example analysis will be carried out using the basic load data derived from 5567 sample load data of a county provided by a power grid company in a certain province of China. This embodiment sets the sampling interval to 30 minutes, and randomly selects 600 samples as experimental samples. The experiment is completed using Matlab2020a on a PC with a CPU main frequency of 3.2GHz and a memory capacity of 16GB. During the verification process, the original load data provided by the data acquisition device has singular values and missing values, and the user load data with more singular values and missing values are re-selected; those with fewer singular values are corrected by a fitting algorithm and the data are normalized to the maximum value.

Taking the 18:30-22:30 period of a certain day in the two-inlet and twenty-four-outlet distribution room of the X cell as an example, the load distribution of the distribution area after aggregation and quantification is analyzed. In this embodiment, the county-specific load aggregation and quantification model based on the sequence operation theory is used to analyze the two users of load A and load B. The obtained load A adjustable capacity curve and load B adjustable capacity curve are shown in Figures 6 and 7, where the load baseline of the control day of load A belongs to the load power consumption mode 1, and the load baseline of the control day of load B belongs to the load power consumption mode 3. The most likely adjustment amount (kW) and the maximum adjustment amount (kW) of the two loads A and B are shown in Table 1:

Table 1

Set the load A to have an adjustable load rate of 95% and the load B to have an adjustable load rate of 85%. Combined with the load profile information, we can obtain the adjustable potential vectors of loads A and B at a certain moment:

R _A (22:00)=[3.03,15.19,95%,0.88]

R _B (22:00)=[1.35,1.19,85%,0.78]

By comparing the two load adjustable potential vectors of load A and load B, it can be seen that the probability of load A being adjustable is significantly higher than that of load B, and the most likely adjustment amount and the maximum adjustment amount are both higher than those of load B. Among them, the maximum adjustment amount of load A is about 5 times the most likely adjustment amount. This is because the most likely adjustment amount does not require the load to adjust the power consumption mode, and is the most easily accepted adjustment method under the load power consumption habits; the most likely adjustment amount of load B is slightly higher than the maximum adjustment amount, indicating that the power consumption behavior of load B at this moment in the near future is basically the same. Combined with the load portrait, it is judged that load B may have rested at this moment. When the adjustment task is released, it is possible to consider lowering the priority of load B.

In this embodiment, it is known that there are 18 loads on the feeder w of the distribution station area. Through load analysis, the adjustable potential vector of the load on feeder w during the regulation period can be obtained. The potential vector of the load at 18:30 is discretized into a simple sequence as shown in Table 2:

Table 2

The potential sequence of the 18 loads on the volume and the feeder can obtain the load adjustable potential probability distribution of feeder w at 18:30 in the regulation period on the regulation day, and the obtained adjustable potential probability distribution of feeder w is shown in Figure 8; the potential sequence of the 24 feeders in the distribution station area can obtain the load adjustable potential distribution of distribution station area l at 18:30 on the regulation day, and the obtained adjustable potential probability distribution fitting result is shown in Figure 9. By comparing Figures 8 and 9, it can be seen in this embodiment that with the increase of the number of users within the calculation range, the overall adjustable potential is expected to increase and tend to the normal distribution. The potential probability distribution at 18:30 is fitted, and the obtained adjustable potential probability distribution fitting result of area I at 18:30 in the regulation period is shown in Figure 10, and the probability density function is shown as follows:

The goodness of fit coefficient is 0.9998, close to 1, indicating that the fitting effect is very good. Therefore, it can be considered that the load adjustable potential of the entire substation at 18:30 during the regulation period obeys the normal distribution. The total load is 1928kW, and the potential expectation is 502kW, which is about 26.0% of the total load. The potential expectation (kW), total load (kW), proportion of regulation, and 95% confidence interval (kW) of this embodiment during the regulation period are shown in Table 3. Table 3 is as follows:

Table 3

This embodiment fits each moment in the regulation period to obtain the fitting result of the probability distribution of adjustable potential after aggregation and quantification of area I as shown in Figure 11. It can be seen from Table 3 and Figure 11 that the adjustable potential of area I in the regulation period is relatively high at around 19:30; it is relatively low at around 22:30, and the adjustment amount accounts for 23%-29%; at 95% confidence level, the daily peak-to-valley difference drops from 1505kW to 981kW at least and to 915kW at most, with significant regulation effect. This embodiment uses a distribution substation with 600 loads for example analysis, fully verifies the effectiveness of the aggregation and quantification model of county-specific loads based on adjustable potential evaluation, and proves that the method proposed in the embodiment of the present invention can accurately grasp the adjustable potential of county-specific loads, assist in the coordinated regulation of new energy, energy storage, and county-specific loads, and provide a model basis for typical load regulation scenarios such as large-scale irrigation period and baking peak season.

An embodiment of the present invention provides a county load aggregation quantification method, which obtains the load power consumption pattern by clustering analysis of the county's historical load curves; determines the load power consumption pattern of the regulation day, and obtains the county's load adjustable amount based on the load power consumption pattern of the regulation day and the load baseline of the regulation day; obtains a single load adjustable potential sequence through discretization based on the obtained load power consumption uncertainty representation and the county's load adjustable amount; obtains the total load adjustable potential probability distribution of the substation feeder according to the single load adjustable potential sequence of the low-voltage feeder within the regulation period of the regulation day; integrates the total load adjustable potential probability distribution of the substation and the real-time data of the substation feeder load into adjustable potential load data, and uses a load aggregation algorithm to iteratively update the adjustable potential load data to obtain the adjustable potential distribution of each load aggregation cluster. The county load aggregation and quantification method proposed in this embodiment adopts a county-specific load adjustable potential model based on power consumption pattern recognition, and at the same time uses the probability distribution of the total load adjustable potential of the substation area to aggregate and quantify the county-specific load, thereby realizing accurate dynamic analysis of the user-side resource adjustment potential, and mastering the spatiotemporal complementary characteristics of the county-specific load and distributed power sources, thereby providing an important basis for subsequent load management and resource allocation optimization, which is of great significance for improving the overall performance of the power grid and promoting the efficient use of energy.

It should be noted that the size of the serial numbers of the above-mentioned processes does not mean 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 application.

In one embodiment, as shown in FIG12 , an embodiment of the present invention provides a county load aggregation and quantification system, the system comprising:

The load pattern analysis module 101 is used to obtain the historical load curve of the county, and obtain the load power consumption pattern by clustering the historical load curve of the county;

The load regulation amount acquisition module 102 is used to determine the load power consumption mode of the regulation day, and obtain the county load adjustable amount based on the load power consumption mode of the regulation day and the load baseline of the regulation day;

The single load potential analysis module 103 is used to calculate the uncertainty characterization of load power consumption, and obtain a single load adjustable potential sequence through discretization based on the uncertainty characterization of load power consumption and the adjustable amount of county load; wherein the uncertainty characterization of load power consumption includes the probability of a single load participating in the regulation task in a certain period and the user adjustable load rate in the regulation period of the regulation day;

The multi-load potential analysis module 104 is used to obtain the probability distribution of the total load adjustable potential of the feeder in the substation area according to the single load adjustable potential sequence of the low-voltage feeder in the regulation period of the regulation day;

The load aggregation quantification module 105 is used to integrate the probability distribution of the adjustable potential of the total load in the substation and the real-time data of the feeder load in the substation collected in advance into adjustable potential load data, and iteratively update the adjustable potential load data using a pre-selected load aggregation algorithm to obtain the adjustable potential distribution of each load aggregation cluster.

In this embodiment, the load adjustment amount acquisition module 102 is specifically used to:

Determine the load baseline for the regulation day according to the daily load curves of several similar days before the regulation day, and calculate the load mode membership for the regulation day according to the load baseline for the regulation day and the typical load curves of different load power consumption modes;

The load power consumption mode corresponding to the maximum load mode membership degree on the control day is selected as the power consumption mode of the control day, and the load adjustable potential of the control day regulation period is calculated according to the load power consumption mode of the control day;

According to the load adjustable potential during the regulation period of the regulation day and the load baseline of the regulation day, the adjustable amount of the county load is calculated; the adjustable amount of the county load includes the most likely adjustable amount of the county load and the maximum adjustable amount of the county load.

In this embodiment, the mathematical expression of the load adjustable potential is:

Where ΔP _m,j (t) represents the load adjustable potential in the regulation period t of the regulation day; It represents the load baseline in the regulation period t of the regulation day; It represents the load curve with the longest Euclidean distance from the load baseline of the regulation day under the same load power consumption mode as the load baseline of the regulation day; m represents the load power consumption mode to which the load baseline of the regulation day belongs; j represents the load power consumption mode after the regulation event occurs; z _φ represents the load curve under the low load power consumption mode within the regulation period t.

In this embodiment, the mathematical expression of the most likely adjustment amount of the county load is:

In the formula, ΔP _α (t) represents the most likely adjustment amount of county load; It represents the load baseline in the regulation period t of the regulation day; It indicates the expected minimum load under the load power consumption mode during the regulation period of the regulation day;

The mathematical expression of the maximum adjustment amount of the county load is:

In the formula, ΔP _β (t) represents the maximum regulation of county load; represents the load baseline within the regulation period t of the regulation day; ^h2 represents the typical load curve under the lowest load power consumption mode among all load power consumption modes.

In this embodiment, the single load potential analysis module is specifically used for:

According to the most likely load regulation amount in the county and the actual load regulation of users on similar days in the historical regulation behavior tags, the actual load regulation rate during the regulation period of similar days is obtained;

According to the actual load regulation rate of the regulation period on similar days and the number of regulation tasks in the same regulation period on similar days, the user adjustable load rate of the regulation period on the regulation day is calculated;

The probability of a single load participating in the regulation task in a certain period of time is calculated according to the response sensitivity coefficient, and the adjustable potential vector is constructed according to the probability of a single load participating in the regulation task in a certain period of time, the user adjustable load rate in the regulation period of the regulation day, and the adjustable amount of the county load.

According to the sequence operation theory, the adjustable potential vector is discretized into a single load adjustable potential sequence.

In this embodiment, the mathematical expression of the load rate that can be adjusted by the user during the adjustment period of the adjustment day is:

in,

Where χ(t) represents the user's adjustable load rate during the regulation period of the regulation day; _Xi represents the actual load regulation rate during the regulation period on similar days; _γi represents the user's successful participation in the demand regulation task; n represents the number of regulation tasks during the same period on similar days; _ΔPi represents the user's actual regulated load on similar days in the historical regulation behavior label; ΔP _α (t) represents the most likely adjustment amount of the user's county load during the regulation period;

The mathematical expression of the probability of a single load participating in the regulation task in a certain period of time is:

Where p(t) represents the probability of a single load participating in the regulation task in a certain period of time; ω represents the sensitivity of the load to participate in the conditional task; τ represents the response sensitivity coefficient; K represents the normalized mean of the historical regulation evaluation score after the user makes effective regulation, where 1≤K≤5.

In this embodiment, the single load adjustable potential sequence should meet the following conditions:

Where ΔP _α (t) represents the most likely adjustment amount of the county load during the adjustment period; χ(t) represents the adjustable load rate of the user during the adjustment period of the regulation day; p(t) represents the probability of a single load participating in the adjustment task during a certain period; A _q represents the adjustable potential sequence of a single load; and _Na represents the sequence length, that is, the theoretical adjustable potential of the load.

In this embodiment, the multi-load potential analysis module is specifically used for:

According to the adjustable potential sequence of a single load of the low-voltage feeder during the regulation period of the regulation day, the probability distribution of the total adjustable potential of the low-voltage feeder load is constructed;

The probability distribution of the total load adjustable potential of each low-voltage feeder in the substation during the regulation period is rolled and calculated to obtain the probability distribution of the total load adjustable potential of the feeder in the substation.

In this embodiment, the method of iteratively updating the adjustable potential load data using a pre-selected load aggregation algorithm to obtain the adjustable potential distribution of each load aggregation cluster specifically includes:

The K-means algorithm is selected as the load aggregation algorithm, and the load aggregation convergence target and load aggregation scale are determined according to the adjustable potential load data;

Several samples are randomly selected from the adjustable potential load data as the initial aggregation centers, and the load aggregation convergence target and load aggregation scale are determined according to the adjustable potential load data. The load aggregation algorithm is used to cluster the adjustable potential load data. After several iterations, the adjustable potential distribution of each load aggregation cluster is obtained.

For the specific limitations of a county load aggregation and quantification system, please refer to the above-mentioned limitations on a county load aggregation and quantification method, which will not be repeated here. A person of ordinary skill in the art will appreciate that the various modules and steps described in conjunction with the embodiments disclosed in this application can be implemented in hardware, software, or a combination of both. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

The embodiment of the present invention provides a county load aggregation quantification system, which performs cluster analysis on the county historical load curve through a load pattern analysis module to obtain the load power consumption pattern; determines the load power consumption pattern of the control day through a load adjustment quantity acquisition module, and obtains the county load adjustable quantity based on the load power consumption pattern of the control day and the load baseline of the control day; the single load potential analysis module obtains a single load adjustable potential sequence through discretization based on the obtained load power consumption uncertainty representation and the county load adjustable quantity; the multi-load potential analysis module obtains the total load adjustable potential probability distribution of the substation feeder according to the single load adjustable potential sequence of the low-voltage feeder within the control period of the control day; the load aggregation quantification module integrates the total load adjustable potential probability distribution of the substation and the real-time data of the substation feeder load into adjustable potential load data, and uses the load aggregation algorithm to iteratively update the adjustable potential load data to obtain the adjustable potential distribution of each load aggregation cluster. The county load aggregation and quantification system proposed in this embodiment adopts a county-specific load adjustable potential model based on power consumption pattern recognition, and uses the probability distribution of the total load adjustable potential of the substation area to aggregate and quantify the county-specific load, thereby realizing accurate dynamic analysis of the user-side resource adjustment potential, and mastering the spatiotemporal complementary characteristics of the county-specific load and distributed power sources, thereby providing an important basis for subsequent load management and resource allocation optimization, which is of great significance for improving the overall performance of the power grid and promoting the efficient use of energy.

The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.

Claims (18)

1. The county-domain load aggregation quantization method is characterized by comprising the following steps of:

Acquiring a county historical load curve, and performing cluster analysis on the county historical load curve to obtain a load electricity utilization mode;

Determining a load electricity consumption mode of a regulating day, and obtaining county load adjustable quantity based on the load electricity consumption mode of the regulating day and a regulating day load baseline;

calculating the load electricity uncertainty characterization, and obtaining a single load adjustable potential sequence through discretization based on the load electricity uncertainty characterization and the county load adjustable quantity; the load electricity uncertainty characterization comprises the probability that a single load participates in a regulation task in a certain period and the user adjustable load rate in a regulation day regulation period;

obtaining total load adjustable potential probability distribution of the feeder line in the transformer area according to a single load adjustable potential sequence of the low-voltage feeder line in a regulation day regulation period;

Integrating the total load adjustable potential probability distribution of the transformer area and the pre-collected real-time data of the feeder line load of the transformer area into adjustable potential load data, and carrying out iterative updating on the adjustable potential load data by utilizing a pre-selected load aggregation algorithm to obtain the adjustable potential distribution of each load aggregation cluster.

2. The county load aggregation quantization method according to claim 1, wherein the step of determining the load electricity consumption mode to which the regulation day belongs and obtaining the county load adjustable amount based on the load electricity consumption mode to which the regulation day belongs and the regulation day load baseline comprises:

Determining a regulating daily load baseline according to daily load curves of a plurality of similar days before the regulating daily, and calculating to obtain the membership of the regulating daily load mode according to the regulating daily load baseline and typical load curves of different load power utilization modes;

selecting a load electricity mode corresponding to the maximum value of the membership degree of the regulating day load mode as an electricity mode to which the regulating day belongs, and calculating the load adjustable potential of the regulating day regulating period according to the load electricity mode to which the regulating day belongs;

calculating to obtain county load adjustable quantity according to the load adjustable potential of the adjustment period of the adjustment day and the adjustment day load base line; the county load adjustable amount includes a county load most likely adjustment amount and a county load maximum adjustment amount.

3. The county-regional load aggregation quantization method of claim 2, wherein the mathematical expression of the load adjustable potential is:

Wherein Δp _m,j (t) represents the load adjustable potential for the regulation day adjustment period t; representing a load baseline over a regulatory day adjustment period t; The load curve with the farthest Euclidean distance from the regulating daily load base line is represented under the condition that the load power consumption mode is the same as the regulating daily load base line; m represents a load electricity utilization mode of regulating and controlling a daily load baseline; j represents a load electricity utilization mode after the occurrence of a regulation event; z _φ represents the load profile in low-load power mode during the conditioning period t.

4. The county load aggregation quantization method according to claim 2, wherein the mathematical expression of the most probable adjustment of county load is:

Wherein Δp _α (t) represents the most likely adjustment amount of county load; representing a load baseline over a regulatory day adjustment period t; representing the lowest load expectation under the load electricity utilization mode of the regulating day in the regulating period;

the mathematical expression of the county load maximum adjustment amount is as follows:

Wherein Δp _β (t) represents the county load maximum adjustment amount; Representing a load baseline over a regulatory day adjustment period t; h ² represents a typical load profile in the lowest load power mode of all load power modes.

5. The county-regional load aggregation quantization method of claim 2, wherein the step of calculating the load electricity uncertainty characterization and obtaining a single load adjustable potential sequence by discretization based on the load electricity uncertainty characterization and the county-regional load adjustable quantity comprises:

According to the most probable adjustment quantity of the county load and the actual adjustment load of the user on the similar day in the history adjustment behavior label, obtaining the actual load adjustment rate of the adjustment period of the similar day;

According to the actual load regulation rate of the regulation time period of the similar day and the regulation task times of the same regulation time period in the similar day, calculating to obtain the user adjustable load rate of the regulation time period of the regulation day;

Calculating the probability of participation in a regulating task of a single load in a certain period according to the response sensitivity coefficient, and constructing an adjustable potential vector according to the probability of participation in the regulating task of the single load in the certain period, the user adjustable load rate of a regulating day regulating period and the county load adjustable quantity;

according to the sequence operation theory, the adjustable potential vector is discretized into a single load adjustable potential sequence.

6. The county-regional load aggregation quantization method according to claim 5, wherein the mathematical expression for regulating the user-adjustable load rate for the day-regulation period is:

wherein,

Wherein χ (t) represents a user-adjustable load rate during the adjustment day adjustment period; chi _i represents the actual load regulation rate for a similar daily regulation period; gamma _i represents the identification of the successful participation of the user in the demand regulation task at this time; n represents the number of simultaneous period regulation tasks in similar days; Δp _i represents the user's actual adjustment load for similar days within the historical adjustment behavior label; Δp _α (t) represents the most likely adjustment amount of county load by the user during the adjustment period;

the mathematical expression of the probability of a single load participating in a regulating task in a certain period is as follows:

Wherein p (t) represents the probability that a single load participates in a regulation task for a certain period of time; omega represents the sensitivity of the load to participate in the conditional task; τ represents a response sensitivity coefficient; k represents the average value of historical adjustment evaluation scoring normalization after effective adjustment by a user, wherein K is more than or equal to 1 and less than or equal to 5.

7. The county-regional load aggregation quantization method according to claim 5, wherein said single load adjustable potential sequence is such that:

Where ΔP _α (t) represents the most likely adjustment amount representation of the county load by the user during the adjustment period; χ (t) represents a user-adjustable load factor for the adjustment day adjustment period; p (t) represents the probability that a single load participates in a regulation task for a certain period of time; a _q represents a single load adjustable potential sequence; n _a represents the sequence length, i.e. the load theory adjustable potential.

8. The county-regional load aggregation quantization method according to claim 1, wherein the step of obtaining the total load adjustable potential probability distribution of the feeder line of the district according to the single load adjustable potential sequence of the low-voltage feeder line in the adjustment time period of the adjustment day comprises:

Constructing total load adjustable potential probability distribution of the low-voltage feeder according to a single load adjustable potential sequence of the low-voltage feeder in a regulation day regulation period;

And rolling and calculating the total load adjustable potential probability distribution of each low-voltage feeder line of the transformer area in the adjustment period to obtain the total load adjustable potential probability distribution of the feeder line of the transformer area.

9. The county-regional load aggregation quantization method of claim 1, wherein the step of iteratively updating the adjustable potential load data using a preselected load aggregation algorithm to obtain an adjustable potential distribution for each load aggregation cluster comprises:

selecting a K-means algorithm as a load aggregation algorithm, and determining a load aggregation convergence target and a load aggregation scale according to the adjustable potential load data;

And randomly selecting a plurality of samples from the adjustable potential load data as an initial aggregation center, determining a load aggregation convergence target and a load aggregation scale according to the adjustable potential load data, clustering the adjustable potential load data by adopting a load aggregation algorithm, and obtaining the adjustable potential distribution of each load aggregation cluster after a plurality of iterations.

10. A county-regional load aggregation quantization system, the system comprising:

The load mode analysis module is used for acquiring a county historical load curve and obtaining a load electricity utilization mode by carrying out cluster analysis on the county historical load curve;

The load adjustment quantity acquisition module is used for determining a load electricity consumption mode of the adjustment day and obtaining county load adjustable quantity based on the load electricity consumption mode of the adjustment day and the adjustment day load baseline;

The single load potential analysis module is used for calculating the load electricity uncertainty characterization, and obtaining a single load adjustable potential sequence through discretization based on the load electricity uncertainty characterization and the county load adjustable quantity; the load electricity uncertainty characterization comprises the probability that a single load participates in a regulation task in a certain period and the user adjustable load rate in a regulation day regulation period;

The multi-load potential analysis module is used for obtaining total load adjustable potential probability distribution of the feeder line in the transformer area according to a single load adjustable potential sequence of the low-voltage feeder line in a regulation day regulation period;

The load aggregation quantization module is used for integrating the total load adjustable potential probability distribution of the platform area and the pre-collected real-time data of the feeder load of the platform area into adjustable potential load data, and carrying out iterative updating on the adjustable potential load data by utilizing a pre-selected load aggregation algorithm to obtain the adjustable potential distribution of each load aggregation cluster.

11. The county-regional load aggregation quantization system according to claim 10, wherein the load adjustment amount acquisition module is specifically configured to:

Determining a regulating daily load baseline according to daily load curves of a plurality of similar days before the regulating daily, and calculating to obtain the membership of the regulating daily load mode according to the regulating daily load baseline and typical load curves of different load power utilization modes;

selecting a load electricity mode corresponding to the maximum value of the membership degree of the regulating day load mode as an electricity mode to which the regulating day belongs, and calculating the load adjustable potential of the regulating day regulating period according to the load electricity mode to which the regulating day belongs;

calculating to obtain county load adjustable quantity according to the load adjustable potential of the adjustment period of the adjustment day and the adjustment day load base line; the county load adjustable amount includes a county load most likely adjustment amount and a county load maximum adjustment amount.

12. The county-regional load aggregation quantization system of claim 11, wherein the mathematical expression for the load adjustable potential is:

Wherein Δp _m,j (t) represents the load adjustable potential for the regulation day adjustment period t; representing a load baseline over a regulatory day adjustment period t; The load curve with the farthest Euclidean distance from the regulating daily load base line is represented under the condition that the load power consumption mode is the same as the regulating daily load base line; m represents a load electricity utilization mode of regulating and controlling a daily load baseline; j represents a load electricity utilization mode after the occurrence of a regulation event; z _φ represents the load profile in low-load power mode during the conditioning period t.

13. The county load aggregation quantization system according to claim 11, wherein the mathematical expression of the most probable adjustment of county load is:

Wherein Δp _α (t) represents the most likely adjustment amount of county load; representing a load baseline over a regulatory day adjustment period t; representing the lowest load expectation under the load electricity utilization mode of the regulating day in the regulating period;

the mathematical expression of the county load maximum adjustment amount is as follows:

Wherein Δp _β (t) represents the county load maximum adjustment amount; Representing a load baseline over a regulatory day adjustment period t; h ² represents a typical load profile in the lowest load power mode of all load power modes.

14. The county-regional load aggregation quantization system of claim 11, wherein the single-load potential analysis module is specifically configured to:

According to the most probable adjustment quantity of the county load and the actual adjustment load of the user on the similar day in the history adjustment behavior label, obtaining the actual load adjustment rate of the adjustment period of the similar day;

According to the actual load regulation rate of the regulation time period of the similar day and the regulation task times of the same regulation time period in the similar day, calculating to obtain the user adjustable load rate of the regulation time period of the regulation day;

Calculating the probability of participation in a regulating task of a single load in a certain period according to the response sensitivity coefficient, and constructing an adjustable potential vector according to the probability of participation in the regulating task of the single load in the certain period, the user adjustable load rate of a regulating day regulating period and the county load adjustable quantity;

according to the sequence operation theory, the adjustable potential vector is discretized into a single load adjustable potential sequence.

15. The county-regional load aggregation quantization system of claim 14, wherein the mathematical expression for regulating the user-adjustable load rate for the daily adjustment period is:

wherein,

Wherein χ (t) represents a user-adjustable load rate during the adjustment day adjustment period; chi _i represents the actual load regulation rate for a similar daily regulation period; gamma _i represents the identification of the successful participation of the user in the demand regulation task at this time; n represents the number of simultaneous period regulation tasks in similar days; Δp _i represents the user's actual adjustment load for similar days within the historical adjustment behavior label; Δp _α (t) represents the most likely adjustment amount of county load by the user during the adjustment period;

the mathematical expression of the probability of a single load participating in a regulating task in a certain period is as follows:

Wherein p (t) represents the probability that a single load participates in a regulation task for a certain period of time; omega represents the sensitivity of the load to participate in the conditional task; τ represents a response sensitivity coefficient; k represents the average value of historical adjustment evaluation scoring normalization after effective adjustment by a user, wherein K is more than or equal to 1 and less than or equal to 5.

16. The county-regional load aggregation quantization system according to claim 14, wherein said single load adjustable potential sequence is such that:

Where ΔP _α (t) represents the most likely adjustment amount representation of the county load by the user during the adjustment period; χ (t) represents a user-adjustable load factor for the adjustment day adjustment period; p (t) represents the probability that a single load participates in a regulation task for a certain period of time; a _q represents a single load adjustable potential sequence; n _a represents the sequence length, i.e. the load theory adjustable potential.

17. The county-regional load aggregation quantization system of claim 10, wherein the multi-load potential analysis module is specifically configured to:

Constructing total load adjustable potential probability distribution of the low-voltage feeder according to a single load adjustable potential sequence of the low-voltage feeder in a regulation day regulation period;

And rolling and calculating the total load adjustable potential probability distribution of each low-voltage feeder line of the transformer area in the adjustment period to obtain the total load adjustable potential probability distribution of the feeder line of the transformer area.

18. The county-regional load aggregation quantization system according to claim 10, wherein the iterative updating of the adjustable potential load data using a preselected load aggregation algorithm results in an adjustable potential distribution for each load aggregation cluster, comprising:

selecting a K-means algorithm as a load aggregation algorithm, and determining a load aggregation convergence target and a load aggregation scale according to the adjustable potential load data;

And randomly selecting a plurality of samples from the adjustable potential load data as an initial aggregation center, determining a load aggregation convergence target and a load aggregation scale according to the adjustable potential load data, clustering the adjustable potential load data by adopting a load aggregation algorithm, and obtaining the adjustable potential distribution of each load aggregation cluster after a plurality of iterations.