Method and system for autonomous operation of power distribution network district with peak regulation as target

Abstract

The invention relates to the technical field of power distribution network district autonomy, in particular to a power distribution network district autonomous operation method and system with peak regulation as the target, and the method comprises the steps: collecting historical load data and meteorological data in a district; predicting the preprocessed data by adopting a long-short-term memory network model to generate a net load inverse correlation curve shape of the regional power distribution network; transformer area resource information is collected, and an autonomous operation transformer area is screened out; based on the transformer area adjusting capability and the load condition, each autonomous operation transformer area uses an optimization algorithm to determine a transformer area plan operation curve; and taking the station area planned operation curve as the target of the station area autonomous operation, implementing the autonomous operation of the station area, adjusting the station area output condition by using a real-time monitoring and feedback mechanism, and minimizing the net load peak-valley difference of the regional power distribution network. Through the inverse correlation curve generated by the algorithm, the transformer area can flexibly adjust the output according to the actual load condition, the peak-valley difference of the net load of the regional power grid is remarkably reduced, and the stability of the power grid is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous operation of distribution network substations, and particularly to a method and system for autonomous operation of a distribution network substation aiming at peak shaving. Background Art

[0002] In modern distribution networks, with the wide access of renewable energy (such as wind energy and solar energy) and the increase of adjustable loads such as electric vehicles and air-conditioning loads, the volatility and uncertainty of the grid load have significantly increased. This has led to problems of load overload during peak hours and energy waste during valley hours in the power grid. Therefore, reducing the peak-valley difference of the net load of the regional power grid, ensuring the safety of the power grid, and improving power supply reliability and economy have become important issues in current distribution network management.

[0003] Traditional load regulation methods mostly rely on static load forecasting and fixed regulation strategies, lacking flexibility and adaptability. In addition, current technical means are difficult to effectively integrate resources between different substations, resulting in unbalanced resource utilization and affecting the overall performance of the power grid. Therefore, there is an urgent need for an autonomous operation method that can achieve dynamic adjustment and intelligent management to optimize the resource allocation of the distribution network, solve the problems of heavy overload of distribution lines and substations, reduce the load peak-valley difference, and improve the flexibility and economy of power supply. Summary of the Invention

[0004] The present invention provides a method and system for autonomous operation of a distribution network substation aiming at peak shaving, which can effectively solve the problems in the background art.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A method for autonomous operation of a distribution network substation aiming at peak shaving, the method comprising:

[0007] Collect historical load data and meteorological data in the substation area, and preprocess the collected data;

[0008] Use a long short-term memory network model to predict the preprocessed data to generate the shape of the anti-correlation curve of the net load of the regional distribution network;

[0009] Collect substation resource information and screen out autonomous operation substations;

[0010] Transmit the shape of the anti-correlation curve of the net load of the regional distribution network to the autonomous operation substations, and based on the adjustment ability and load conditions of the substations, each autonomous operation substation uses an optimization algorithm to determine the planned operation curve of the substation;

[0011] Take the planned operation curve of the substation area as the goal of autonomous operation of the substation area, implement the autonomous operation of the substation area, and use the real-time monitoring and feedback mechanism to adjust the power output of the substation area to minimize the peak-valley difference of the net load of the regional distribution network.

[0012] Further, it includes: performing clustering analysis on the daily load curve of the power system, calculating the covariance between variables, and defining the curve shape distance between two variables. The formula is:

[0013] D(X,Y) = 1 - ρ(X,Y);

[0014] Where D(X,Y) is the curve shape distance; ρ(X,Y) is the covariance between variables.

[0015] Further, the generation of the anti-correlated curve shape of the net load of the regional distribution network includes: using the long short-term memory network model to generate the net load curve X of the regional distribution network, taking the curve -X that is symmetric about the x-axis of the net load curve X of the regional distribution network, and performing operations of shrinking and upward translation on the curve -X to form the anti-correlated curve -aX + b of the net load of the regional distribution network.

[0016] Further, it also includes: defining the size of the peak-valley difference of the net load curve of the regional distribution network, and using the peak-valley difference rate of the net load curve = (peak value of the net load curve - valley value of the net load curve) / peak value of the net load curve. The formula is:

[0017]

[0018] Where, x max is the peak value of the net load curve; x min is the valley value of the net load curve;

[0019] The peak-valley difference rate after superimposing the anti-correlated curve of the net load of the regional distribution network is calculated by the formula:

[0020]

[0021] Where a is the scaling coefficient, 0 < a < 1; b is the translation coefficient, b > 0.

[0022] Further, based on the adjustment ability and load situation of the substation area, each autonomous operation substation area uses an optimization algorithm to determine the planned operation curve of the substation area, including:

[0023] Through the collected substation area resource information, predict the substation area load and distributed new energy output, and analyze the adjustable resources;

[0024] According to the analysis results, calculate the peak value and valley value of the net load, determine the scaling coefficient a and translation coefficient b of the anti-correlated curve of the net load of the regional distribution network, and obtain the anti-correlated curve of the net load of the regional distribution network;

[0025] Taking the net load anti-correlation curve of the regional distribution network as a comparison benchmark, a planned operation curve of the substation area with the smallest curve shape distance from the net load anti-correlation curve of the regional distribution network is established through an optimization algorithm.

[0026] Further, it includes: the objective function of the optimization algorithm is:

[0027] obj = minD(A', B);

[0028] Where D is the curve shape distance; A' is the planned operation curve of the substation area; B is the net load anti-correlation curve of the regional distribution network.

[0029] Further, it also includes that the constraint of the optimization algorithm is the output and safety constraints of various resources in the substation area.

[0030] Further, it includes: the real-time monitoring and feedback mechanism is realized through Internet of Things devices, and the Internet of Things devices include sensors, data acquisition modules and control systems.

[0031] Further, it also includes: judging the peak-valley difference rate of the net load curve after the autonomy of the substation area through the description of the peak-valley difference of the net load curve of the regional distribution network. If no thinning points are taken to control the substation area resources, the net load curve without taking the means of substation area autonomy is obtained according to the actual sizes of distributed power sources and loads, and at the same time, the peak-valley difference rate of the net load curve is obtained.

[0032] A distribution network substation area autonomous operation system aiming at peak regulation, the system includes:

[0033] A data acquisition module, which collects historical load data and meteorological data in the substation area and preprocesses the collected data;

[0034] A curve generation module, which predicts the preprocessed data using a long short-term memory network model to generate the shape of the net load anti-correlation curve of the regional distribution network;

[0035] A substation area screening module, which collects substation area resource information and screens out autonomous operation substation areas;

[0036] A plan formulation module, which transmits the shape of the net load anti-correlation curve of the regional distribution network to the autonomous operation substation area. Based on the adjustment ability and load conditions of the substation area, each autonomous operation substation area uses an optimization algorithm to determine the planned operation curve of the substation area;

[0037] A substation area peak regulation module, which takes the planned operation curve of the substation area as the goal of the autonomous operation of the substation area, implements the autonomous operation of the substation area, and uses the real-time monitoring and feedback mechanism to adjust the output situation of the substation area to minimize the peak-valley difference of the net load of the regional distribution network.

[0038] Through the technical solution of the present invention, the following technical effects can be achieved:

[0039] It effectively solves the problem that it is difficult to integrate the resources of the distribution network substation area. The anti-correlation curve generated by the algorithm of the present invention enables the substation area to flexibly adjust the output according to the actual load situation, significantly reducing the peak-valley difference of the net load of the regional power grid and improving the stability of the power grid; the optimization algorithm can ensure that each substation area flexibly adjusts the load according to factors such as its own capacity and market electricity price, realizing the optimal allocation of resources and improving the overall operation efficiency; the real-time feedback mechanism combined with the Internet of Things technology can dynamically monitor and adjust the resource output, enabling the system to operate efficiently under changing load and environmental conditions.

[0040] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specifically gives the specific implementation manners of this application. Description of the Drawings

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0042] Figure 1 It is a schematic flowchart of the autonomous operation method of the distribution network substation area with peak regulation as the goal;

[0043] Figure 2 It is a network topology diagram of an embodiment of the autonomous operation method of the distribution network substation area with peak regulation as the goal;

[0044] Figure 3 It is a schematic diagram of the net load mirror image of substation area 1 of an embodiment of the autonomous operation method of the distribution network substation area with peak regulation as the goal;

[0045] Figure 4 It is a schematic diagram of the net load mirror image of the autonomous operation method of the distribution network substation area with peak regulation as the goal;

[0046] Figure 5 It is a schematic diagram of the comparison of the net load curves of the distribution network substation area of the autonomous operation method with peak regulation as the goal. Detailed Embodiments

[0047] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0049] Embodiment 1

[0050] As Figure 1 shown, a method for autonomous operation of a distribution network substation area aiming at peak regulation includes:

[0051] S1: Collect historical load data and meteorological data in the substation area, and preprocess the collected data;

[0052] S2: Use the long short-term memory network model to predict the preprocessed data to generate the shape of the anti-correlation curve of the net load of the regional distribution network;

[0053] Specifically, to ensure the quality and accuracy of the collected data, the data can be preprocessed, abnormal data can be checked and processed, and the data itself can be used as a feature to train the long short-term memory network model to generate the net load curve of the regional distribution network. According to this curve, a shape of the anti-correlation curve of the net load curve is formed. When this anti-correlation curve shape is superimposed on the regional net load curve, it can make the shape of the regional net load curve approach a straight line, thereby reducing the peak-valley difference. The net load curve is subject to power balance constraints.

[0054] S3: Collect substation area resource information and screen out the substation areas for autonomous operation;

[0055] S4: Transmit the shape of the anti-correlation curve of the net load of the regional distribution network to the substation areas for autonomous operation. Based on the adjustment ability and load conditions of the substation areas, each substation area for autonomous operation uses an optimization algorithm to determine the planned operation curve of the substation area;

[0056] In this embodiment, the substation areas with adjustable resources and strong adjustable ability are selected as the substation areas for autonomous operation, and the predicted shape of the anti-correlation curve of the net load of the regional power grid is transmitted to each substation area for autonomous operation. Each substation area can use an optimization algorithm to determine a suitable planned operation curve according to its own adjustment ability and load conditions, and use this planned operation curve as the goal of autonomous operation.

[0057] S5: Use the planned operation curve of the substation area as the goal of autonomous operation of the substation area, implement the autonomous operation of the substation area, and use the real-time monitoring and feedback mechanism to adjust the output of the substation area to minimize the peak-valley difference of the net load of the regional distribution network.

[0058] Specifically, during the real-time operation process, the real-time monitoring and feedback mechanism can be used to collect the load data and resource output data of the substation area in real time through Internet of Things devices, and continuously adjust the output of the resources on the substation side, so that the net load curve of the substation area approaches the autonomous operation curve, thereby achieving the goal of reducing the peak-valley difference of the net load of the regional power grid.

[0059] The present invention effectively solves the problem that it is difficult to integrate the resources of the distribution network substation area. The anti-correlation curve generated by the algorithm of the present invention enables the substation area to flexibly adjust the output according to the actual load situation, significantly reducing the peak-valley difference of the net load of the regional power grid and improving the stability of the power grid; the optimization algorithm can ensure that each substation area flexibly adjusts the load according to factors such as its own capabilities and market electricity prices, realizing the optimal allocation of resources and improving the overall operation efficiency; the real-time feedback mechanism combined with Internet of Things technology can dynamically monitor and adjust the resource output, enabling the system to operate efficiently under changing load and environmental conditions.

[0060] Furthermore, it includes: performing clustering analysis on the daily load curve of the power system, calculating the covariance between variables, and defining the curve shape distance between two variables. The formula is:

[0061] D(X,Y) = 1 - ρ(X,Y);

[0062] where D(X, Y) is the curve shape distance; ρ(X, Y) is the covariance between variables.

[0063] As a preference of this embodiment, perform clustering analysis on the daily load curve of the power system and calculate the covariance between variables. The formula is:

[0064]

[0065] Derive the curve shape distance of the curve shapes of two variables through the covariance between the two variables, which has the following characteristics:

[0066] (1) Symmetry D(X,Y) = D(Y,X)

[0067] (2) Non-negativity D(X,Y) ≥ 0

[0068] (3) Self-distance is zero D(X,X) = 0

[0069] (4) If the distance is zero, the two curve shapes are the same

[0070] (5) Translation and scaling invariance D((aX + b),Y) = D(X,Y)

[0071] The similarity between curves can be determined by the above formula, which can serve as the basis for the correlation analysis of the net load curves.

[0072] Furthermore, the shape of the anti-correlation curve of the regional distribution network net load includes: generating the regional distribution network net load curve X using the long short-term memory network model, taking the curve -X that is symmetric about the x-axis for the regional distribution network net load curve X, and performing operations of shrinking and upward translation on the curve -X to form the regional distribution network net load anti-correlation curve -AX + b.

[0073] Specifically, directly take the curve -X that is symmetric about the x-axis for a certain net load curve X, and then form the anti-correlation net load curve -aX + b through shrinking and upward translation operations. This curve has the opposite direction to the regional net load curve, which plays the role of smoothing the load change and reducing the peak-valley difference. By superimposing this curve with the regional net load curve, the overall load can be made more stable and the load volatility can be reduced. For the specific curve diagram, refer to Figure 4 shown.

[0074] In order to describe the magnitude of the peak-valley difference of the regional distribution network net load curve, a mathematical description method can be defined, including: defining the magnitude of the peak-valley difference of the regional distribution network net load curve, using the peak-valley difference rate of the net load curve = (peak value of the net load curve - valley value of the net load curve) / peak value of the net load curve, and the formula is:

[0075]

[0076] where, x max is the peak value of the net load curve; x min is the valley value of the net load curve;

[0077] The peak-valley difference rate after superimposing the regional distribution network net load anti-correlation curve, the formula is:

[0078]

[0079] where, a is the scaling coefficient, 0 < a < 1; b is the translation coefficient, b > 0.

[0080] Specifically, as Figure 3 shown, the curve form generated by mirroring the net load curve. This mirror curve reflects the symmetric relationship of the net load on the time axis, which helps to analyze the trend and characteristics of the load change and provides a reference for the autonomous operation strategy of the substation area.

[0081] The present invention conducts a day-ahead prediction of the net load for a certain area through historical data and meteorological information data, obtains the predicted net load curve A, and obtains the net load inverse correlation curve -A through the above mirror method. The shape of this curve is used as a reference basis, and subsequent specific curve values are determined through the evaluation of the adjustment ability of the distribution transformer area.

[0082] In order to determine the specific values of the net load inverse correlation curve of the regional distribution network, based on the adjustment ability of the distribution transformer area and the load situation, each autonomous operation distribution transformer area uses an optimization algorithm to determine the planned operation curve of the distribution transformer area, including:

[0083] A1: Through the collected distribution transformer area resource information, predict the load of the distribution transformer area, the output of distributed new energy, and analyze the adjustable resources;

[0084] A2: According to the analysis results, calculate the peak value and valley value of the net load, determine the scaling coefficient a and translation coefficient b of the net load inverse correlation curve of the regional distribution network, and obtain the net load inverse correlation curve of the regional distribution network;

[0085] Specifically, the adjustable resources include energy storage, adjustable load, distributed photovoltaic, etc. According to the characteristics of the adjustable resources, their adjustment capabilities under different load conditions can be analyzed. For example, the energy storage system can provide power support during peak load periods. This step uses the resource information of the distribution transformer area and the predictions of the load and the output of renewable energy to calculate the peak and valley values of the net load, that is, the maximum and minimum values that the net load of the distribution transformer area can be adjusted. Combining the adjustment ability of the distribution transformer area, determine the scaling coefficient and translation coefficient of the inverse correlation net load curve, and finally obtain the target inverse correlation curve to help adjust the load of the distribution transformer area, reduce the peak-valley difference, and optimize the grid load.

[0086] A3: Use the net load inverse correlation curve of the regional distribution network as a comparison benchmark, and establish a planned operation curve of the distribution transformer area with the smallest curve shape distance from the net load inverse correlation curve of the regional distribution network through an optimization algorithm.

[0087] Furthermore, the objective function of the optimization algorithm is:

[0088] obj = minD(A', B);

[0089] where D is the curve shape distance; A' is the planned operation curve of the distribution transformer area; B is the net load inverse correlation curve of the regional distribution network.

[0090] As a preference of this embodiment, according to the determined net load inverse correlation curve of the distribution transformer area as the comparison benchmark B = -aA + b, establish an autonomous operation curve A' of the distribution transformer area with the smallest curve shape distance through an optimized method, which can take minimizing the error between the actual load curve after autonomy and the target reference curve as the core objective, and its mathematical expression is:

[0091]

[0092] Among them, Y actual (t) represents the actual net load at a certain time t after autonomous operation; Y tar It represents the target reference load curve, which is usually a net load anti-correlation curve generated by anti-correlation operation combined with the load resource situation in the substation. By minimizing the sum of squares of errors, it ensures that the load curve after autonomy fits the target curve as closely as possible within the entire time range, reflecting the integrity of the optimization results.

[0093] On the basis of the above-mentioned embodiment, the optimization algorithm is also constrained by the output and safety constraints of various resources in the substation.

[0094] Specifically, the optimization process needs to meet the following constraints to ensure the feasibility of autonomous operation and grid security:

[0095] (1) Power balance constraints:

[0096] P pv =P load -P ess +P dis -P IDR ;

[0097] Where: P pv For photovoltaic output; P load is the load output; P ess is the energy storage charging power; P dis is the discharge power of energy storage; P IDR For adjustable load output.

[0098] (2) Load constraints:

[0099] After autonomous operation, the load must fluctuate within the permissible range:

[0100]

[0101] Where: P min and P max They are the upper and lower limits of the system operating load, respectively, to ensure that the load curve does not exceed the system carrying capacity.

[0102] (3) Energy storage constraints:

[0103]

[0104] Where: E i,t is the charge of the energy storage connected to node i at time t; and are the energy storage charging efficiency and discharging efficiency of access node i, respectively; and are the charging power and discharging power of the energy storage connected to node i at time t; and are the charge and discharge status indicators of the energy storage connected to node i at time t, which are 0 or 1; Δt is the time step, which is 1 hour; E i,0 and E i,T are the energy storage charges of access node i at the initial time and the end time respectively; E i,min They are the lower limit of energy storage charge at node i; are the rated power and rated capacity of the energy storage at access node i respectively.

[0105] (4) Adjustable load constraints:

[0106] The adjustable load takes into account the translatable load and should meet the following requirements:

[0107] P IDR,min ≤P IDR ≤P IDR，max ;

[0108] Where: P IDR,min is the minimum output of the load that can be translated; P IDR，max It is the maximum value of the load output that can be translated.

[0109] As a preferred embodiment of this embodiment, the real-time monitoring and feedback mechanism is implemented through an Internet of Things device, which includes a sensor, a data acquisition module and a control system.

[0110] Furthermore, it also includes: judging the peak-to-valley difference rate of the net load curve after substation autonomy through the peak-to-valley difference description of the regional distribution network net load curve; if no thin points are adopted to control substation resources, the net load curve without taking substation autonomy measures is obtained according to the actual size of the distributed power source and the load, and the peak-to-valley difference rate of the net load curve is obtained at the same time.

[0111] Specifically, Figure 2 The example topology diagram is used as an example for simulation. Figure 2 There are 10 nodes in total, each node represents a substation. Some substations cannot be autonomous because they do not contain adjustable resources or the adjustable resources cannot participate in communication. Select node 5 in the example for substation autonomy, where the maximum output of the adjustable load is 20kW, the maximum capacity of the energy storage is 50kW, and the maximum output of the photovoltaic is 200kW. First, by processing the historical load data and meteorological data, the net load curve is generated by the mirror operation, as shown in the following figure. Figure 3 As shown; then consider the total amount of available regulation resources in node 5, perform appropriate translation and scaling operations on the mirrored net load curve, and generate an anti-correlated net load curve, as shown Figure 4As shown; it can be seen that at most time points, the system autonomy results are the same as expected. There are also some time points such as 9 o'clock and 21 o'clock, where the trend of the curve is different from the net load curve after mirroring. This is because the available regulation resources in the substation have reached the upper limit.

[0112] Last reference Figure 5 After determining the appropriate planned operation curve, the optimized net load curve and the shape distance of the comparison benchmark can be obtained according to the result. This will reflect the adjustable capacity of the substation's own resources, that is, whether the substation has the ability to adjust the output according to the shape of the anti-correlated net load curve, and select the substation with stronger capacity as the autonomous operation substation. The autonomous operation substation uses the results of the day-ahead optimization as the reference curve benchmark for intraday operation, but due to the deviation of the prediction, the actual operation results will deviate from the results of the day-ahead optimization; the peak-to-valley difference rate of the net load curve after the substation's autonomy is judged by describing the peak-to-valley difference of the net load curve. It can be obtained that the peak-to-valley rate without the substation's autonomous operation is 8.9%, while the peak-to-valley rate after the substation's autonomy is 2.8%, indicating that the load curve is smoother and the peak-to-valley difference is smaller after autonomy.

[0113] Embodiment 2:

[0114] A distribution network area autonomous operation system for peak load regulation, the system comprising:

[0115] The data collection module collects historical load data and meteorological data in the substation area and pre-processes the collected data;

[0116] The curve generation module uses the long short-term memory network model to predict the preprocessed data and generate the anti-correlation curve shape of the net load of the regional distribution network;

[0117] The station area screening module collects station area resource information and screens out autonomous operation stations;

[0118] The planning module transmits the shape of the anti-correlation curve of the net load of the regional distribution network to the autonomous operation area. Based on the regulation capacity and load conditions of the area, each autonomous operation area uses an optimization algorithm to determine the planned operation curve of the area;

[0119] The substation peak-shaving module takes the substation planned operation curve as the goal of substation autonomous operation, implements substation autonomous operation, and uses real-time monitoring and feedback mechanisms to adjust the substation output to minimize the peak-to-valley difference in net load of the regional distribution network.

[0120] The above-mentioned adjustment system in the present invention can effectively realize the autonomous operation method of the distribution network substation with the goal of peak load regulation. The technical effects that can be achieved are as described in the above-mentioned embodiments and will not be repeated here.

[0121] Although the present application has been described in conjunction with specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined herein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for autonomous operation of a distribution network area with the goal of peak load regulation, characterized in that: The method includes: Collecting historical load data and meteorological data in the substation area and preprocessing the collected data; Using a long short-term memory network model to predict the preprocessed data and generating the shape of the inverse correlation curve of the net load of the regional distribution network; Collecting substation area resource information and screening out self-governing operation substations; Transmitting the shape of the inverse correlation curve of the net load of the regional distribution network to the self-governing operation substations. Based on the adjustment ability and load conditions of the substations, each self-governing operation substation uses an optimization algorithm to determine the planned operation curve of the substation; Taking the planned operation curve of the substation as the goal of the self-governing operation of the substation, implementing the self-governing operation of the substation, and using a real-time monitoring and feedback mechanism to adjust the output of the substation to minimize the peak-valley difference of the net load of the regional distribution network.

2. The method for autonomous operation of a distribution network area with the goal of peak load regulation according to claim 1, characterized in that: Including: Performing clustering analysis on the daily load curve of the power system, calculating the covariance between variables, and defining the curve shape distance between two variables. The formula is: D(X,Y) = 1 - ρ(X,Y); Where D(X,Y) is the curve shape distance; ρ(X,Y) is the covariance between variables.

3. The method for autonomous operation of a distribution network area with the goal of peak load regulation according to claim 1, characterized in that: The generating of the shape of the inverse correlation curve of the net load of the regional distribution network includes: using the long short-term memory network model to generate the net load curve X of the regional distribution network, taking the curve -X that is symmetric about the x-axis of the net load curve X of the regional distribution network, and performing operations of shrinking and upward translation on the curve -X to form the inverse correlation curve -aX + b of the net load of the regional distribution network.

4. The method for autonomous operation of a distribution network area with the goal of peak load regulation according to claim 1, characterized in that: Also including: Defining the size of the peak-valley difference of the net load curve of the regional distribution network, and using the peak-valley difference rate of the net load curve = (peak value of the net load curve - valley value of the net load curve) / peak value of the net load curve. The formula is: Among them, x max is the peak value of the net load curve; x min is the valley value of the net load curve; The peak-valley difference rate after superimposing the inverse correlation curve of the net load of the regional distribution network. The formula is: Where a is the scaling coefficient, 0 < a < 1; b is the translation coefficient, b > 0.

5. The method for autonomous operation of a distribution network area with the goal of peak load regulation according to claim 3, characterized in that: Based on the adjustment ability and load conditions of the substations, each self-governing operation substation uses an optimization algorithm to determine the planned operation curve of the substation, including: Through the collected substation area resource information, predicting the substation load and distributed new energy output, and analyzing the adjustable resources; According to the analysis results, calculating the peak value and valley value of the net load, determining the scaling coefficient a and translation coefficient b of the inverse correlation curve of the net load of the regional distribution network, and obtaining the inverse correlation curve of the net load of the regional distribution network; Taking the inverse correlation curve of the net load of the regional distribution network as the comparison benchmark, and establishing a planned operation curve of the substation with the minimum curve shape distance from the inverse correlation curve of the net load of the regional distribution network through an optimization algorithm.

6. A method for autonomous operation of a distribution network area with the goal of peak load regulation according to claim 5, characterized in that: Including: The objective function of the optimization algorithm is: obj = minD(A',B); Where D is the curve shape distance; A’ is the planned operation curve of the substation; B is the inverse correlation curve of the net load of the regional distribution network.

7. A method for autonomous operation of a distribution network area with the goal of peak load regulation according to claim 6, characterized in that: The constraints of the optimization algorithm also include the output and safety constraints of various resources in the substation area.

8. The method for autonomous operation of a distribution network area with the goal of peak load regulation according to claim 1, characterized in that: Including: The real-time monitoring and feedback mechanism is implemented through Internet of Things devices, and the Internet of Things devices include sensors, data acquisition modules and control systems.

9. The method for autonomous operation of a distribution network area with the goal of peak load regulation according to claim 1, characterized in that: Also including: The peak-to-valley difference rate of the net load curve after substation autonomy is determined by describing the peak-to-valley difference of the net load curve of the regional distribution network. If no thin points are taken to control substation resources, the net load curve without substation autonomy is obtained based on the actual size of the distributed power source and the load, and the peak-to-valley difference rate of the net load curve is also obtained.

10. A distribution network area autonomous operation system for peak load regulation, characterized in that: The system comprises: The data collection module collects historical load data and meteorological data in the substation area and pre-processes the collected data; The curve generation module uses the long short-term memory network model to predict the preprocessed data and generate the anti-correlation curve shape of the net load of the regional distribution network; The station area screening module collects station area resource information and screens out autonomous operation stations; A planning module transmits the shape of the net load anti-correlation curve of the regional distribution network to the autonomous operation area, and based on the regulation capacity and load conditions of the area, each autonomous operation area uses an optimization algorithm to determine the planned operation curve of the area; The substation peak-shaving module takes the substation planned operation curve as the target of the substation autonomous operation, implements the substation autonomous operation, and uses real-time monitoring and feedback mechanism to adjust the substation output to minimize the peak-to-valley difference of the net load of the regional distribution network.

Images