Distribution network balance optimization method considering distributed resource consumption control

Abstract

The invention discloses a distribution network balance optimization method considering distributed resource consumption control, and the method comprises the following steps: S1, predicting the total charging load of an electric vehicle, and obtaining a prediction result of the load density in a distribution network; s2, predicting the output of the distributed photovoltaic energy to obtain a power density prediction result in the distribution network; s3, performing preliminary approximate division on the power supply area; s4, determining a load center and a power center based on a clustering analysis method; s5, constructing a power distribution network power supply area division optimization model; s6, solving the power distribution network power supply area division optimization model based on a minimum weight matching method, and determining the power supply range of each transformer substation; according to the method, distributed resources are considered, the load and the power supply density in the distribution network are predicted, the power supply range is determined through clustering analysis and construction of the power distribution network power supply area division optimization model, local and nearby consumption of the distributed power supply is achieved, and the balance of the load and the resource of the partition is ensured.

Description

Technical Field

[0001] This invention belongs to the field of distribution network balancing optimization technology, and specifically relates to a distribution network balancing optimization method that takes into account distributed resource consumption control. Background Technology

[0002] In the future, the penetration rate of distributed resources, including electric vehicles and distributed photovoltaics, will increase significantly in the distribution network. The distribution network will also transform from a traditional single role of power distribution into a new type of power exchange system that integrates power collection, transmission, storage and distribution. The rapid development of distributed resources is both a challenge and an opportunity for the distribution network.

[0003] Due to the significant temporal, decentralized, and fluctuating characteristics of distributed photovoltaic (PV) power output, it has a certain coupling relationship with distribution network loads and other new loads in terms of time and space. Its large-scale grid connection poses more complex challenges to the new power system. At the same time, with the gradual increase in the scale of electric vehicles, the uncertainty brought about by electric vehicle loads has also had a relatively negative impact on the stable operation of the distribution network. Therefore, considering the flow of distributed energy, optimizing the division of the distribution network, realizing the local consumption of distributed power sources, and ensuring the balance of regional loads and resources are very important for distribution networks with high-penetration distributed resources.

[0004] Therefore, in order to solve the above problems, it is necessary to develop a distribution network balance optimization method that takes into account the control of distributed resource consumption. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a distribution network balance optimization method that takes into account the control of distributed resource consumption. It takes into account distributed resources, predicts the load and power density in the distribution network, and determines the power supply range by cluster analysis and constructing an optimization model for the division of power supply areas in the distribution network, so as to realize the local consumption of distributed power sources and ensure the balance of load and resources in the region.

[0006] The objective of this invention is achieved as follows: a distribution network balancing optimization method considering distributed resource absorption control, comprising the following steps:

[0007] S1. Based on the demand for electric vehicles in the area to be planned in the distribution network, combined with the typical charging load curve of electric vehicles and the overall scale of electric vehicles, the total charging load of electric vehicles in the area to be planned is predicted, and the load density prediction result in the distribution network is obtained.

[0008] S2. Based on the seasonal variation of the installed area of ​​the distributed photovoltaic system and the solar irradiance of the region, a spatiotemporal multi-scale output model of distributed photovoltaic is constructed to predict the output of distributed photovoltaic energy and obtain the power density prediction results in the distribution network.

[0009] S3. Based on the load density and power density prediction results, the power supply area is initially approximated and divided into input power supply area and output power supply area;

[0010] S4. Based on cluster analysis, determine the load center and power center in each power supply area after preliminary approximate division, and allocate the power and load in each power supply area to the nearest clustered load and power group center.

[0011] S5. Using the minimum sum of distances between load centers and power centers within the power supply area of ​​each substation as the objective function, and the maximum allowable load rate and power supply radius as constraints, construct an optimization model for the division of power supply areas in the distribution network.

[0012] S6. Solve the power supply area division optimization model of the distribution network based on the minimum weight matching method to determine the power supply range of each substation.

[0013] Furthermore, in step S1, the Monte Carlo method is used to calculate typical charging load curves for various types of electric vehicles in the area to be planned on different charging facilities.

[0014] Furthermore, the total charging load of electric vehicles in the area to be planned in step S1 is expressed as: In the formula: express Total regional charging load at any given time; Represents a set of electric vehicle types; Indicates the area to be planned The projected size of this type of vehicle; This represents the typical charging load for an electric vehicle. In the formula: express Car-like Typical charging load values ​​for a given period of time This represents a set of charging facility types. express similar cars The proportion of people charging at similar charging facilities. express similar vehicles at charging facilities On Typical charging load during the time period.

[0015] Furthermore, the electric vehicle charging scale in the electric vehicle demand in step S1 is expressed as: In the formula: , These represent the load allocation weight and area allocation weight, respectively. and They represent the first The typical load value and land area of ​​each plot, This indicates the total number of plots of land in this type of functional zone within the area to be planned; Represents a set of electric vehicle types; Indicates the first A collection of charging facility types for each plot of land; express similar cars The number of devices charging at these types of charging facilities. , express similar cars The proportion of people charging at similar charging facilities. express The predicted size of the vehicle type; then the land parcel of The charging load during a given period is expressed as follows: .

[0016] Furthermore, the spatiotemporal multi-scale output model of distributed photovoltaic power in step S2 is expressed as follows: In the formula: express Regional distributed photovoltaic in Efforts made at all times; Indicates photoelectric conversion efficiency; This indicates the actual installed area of ​​distributed photovoltaic power generation. express Solar irradiance per unit area at any given time.

[0017] Furthermore, in step S2, the actual installed area of ​​distributed photovoltaic power... In the formula: express The available installation area for distributed photovoltaic power within the region. In the formula, express Total above-ground building area within the area express The area's floor area ratio, express Average number of floors in buildings within the area express The rate at which rooftop photovoltaic installations are feasible in the area;

[0018] This represents the floor space occupied by a single photovoltaic module. If two photovoltaic panels are considered as a single photovoltaic module and arranged in an array, then... In the formula, Indicates the minimum distance between photovoltaic arrays. This indicates the optimal tilt angle for installing photovoltaic panels. and These represent the length and width of a single photovoltaic panel, respectively. This indicates the area of ​​a single photovoltaic module; Indicates not greater than The largest integer.

[0019] Furthermore, the preliminary approximate division of the power supply area in step S3 is determined based on the load and power balance results. At that time, it is an input-type power supply area, when At that time, it is an output power supply area, in which, Indicates the grid supply load. This indicates grid-connected power supply.

[0020] Furthermore, step S4 employs an improved K-means clustering method for cluster analysis, specifically including the following steps: ① Determining the number of target clustering areas. ; In the formula: This represents the floor function. This represents the total load / power supply within the area to be divided in the distribution network. This indicates the target capacity of the zoned substations. ① Represent the power factor of the substation; ② Divide the area to be divided into A square redundant grid is used, and the center of each grid is selected as the initial load / power supply center. Choose a value greater than the required precision. A certain value; ③ Distribute all loads / power supplies to the nearest available location. ④ Eliminate the grid with the least load / power supply and allocate its load / power supply to the nearest grid. The number of redundant grids is [number missing]. Subtract 1; ⑤ Repeat step ④ until... ,get ⑥ Calculate the weighted distance from each load / power source to its center: In the formula, Indicates load / power supply To the load / power center European distance, , Indicates load / power center Distance weighting factor, In the formula, This represents the total load / power supply within the area to be divided in the distribution network. Indicates load / power center The total load / power supply obtained in the previous partitioning iteration. Indicates the weighting factor adjustment coefficient; ⑦ Distribute all loads / power sources according to the principle of minimum weighted distance to 8. Obtain the load / power center for each zone, and use the center point of each zone's load / power as the set of load / power center points for the next iteration; 9. Repeat steps 6-7 until the load / power center and zone range no longer change or the change is less than a certain precision; 10. Obtain the clustering results.

[0021] Furthermore, the objective function in step S5 is expressed as: In the formula: A set representing substation numbers; Indicates substation Load centers within service area and power center The distance between them.

[0022] Furthermore, the constraint conditions in step S5 are expressed as follows: In the formula: These represent the substation's load rate and maximum allowable load rate, respectively. These represent the power supply radius of the substation outgoing lines and their maximum allowable value, respectively.

[0023] Due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0024] (1) By predicting the load and power output of typical distributed resources in the distribution network, and making preliminary approximate divisions of the power supply area of ​​the distribution network based on the prediction results, power load matching is carried out to ensure balanced power load distribution, improve the utilization rate of distributed power sources, and lay the foundation for subsequent division.

[0025] (2) Based on the cluster analysis algorithm, the load center and power center in the region after the initial division are determined, which facilitates the consideration of energy flow, ensures source-load matching, and realizes the local consumption of distributed resources;

[0026] (3) By constructing an optimization model for the division of power supply areas in the distribution network and using the minimum weight matching method to solve the model, the power supply range is determined, distributed resources are consumed nearby, the balance of load and resources in the region is improved, and the safety, reliability and stability of the distribution network operation are enhanced. Attached Figure Description

[0027] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0028] The technical solution of the present invention will be further described in detail below through embodiments and in conjunction with the accompanying drawings.

[0029] like Figure 1 As shown, a distribution network balancing optimization method considering distributed resource consumption control includes the following steps:

[0030] S1. Based on the demand for electric vehicles in the area to be planned in the distribution network, combined with the typical charging load curve of electric vehicles and the overall scale of electric vehicles, the total charging load of electric vehicles in the area to be planned is predicted, and the load density prediction result in the distribution network is obtained.

[0031] Preferably, in step S1, the Monte Carlo method is used to calculate typical charging load curves for various types of electric vehicles in the area to be planned on different charging facilities.

[0032] Preferably, the total charging load of electric vehicles in the area to be planned in step S1 is expressed as: In the formula: express Total regional charging load at any given time; Represents a set of electric vehicle types; Indicates the area to be planned The projected size of this type of vehicle; This represents the typical charging load for an electric vehicle. In the formula: express Car-like Typical charging load values ​​for a given period of time This represents a set of charging facility types. express similar cars The proportion of people charging at similar charging facilities. express similar vehicles at charging facilities On Typical charging load during the time period.

[0033] Preferably, the electric vehicle charging scale in the electric vehicle demand in step S1 is expressed as: In the formula: , These represent the load allocation weight and area allocation weight, respectively. and They represent the first The typical load value and land area of ​​each plot, This indicates the total number of plots of land in this type of functional zone within the area to be planned; Represents a set of electric vehicle types; Indicates the first A collection of charging facility types for each plot of land; express similar cars The number of devices charging at these types of charging facilities. , express similar cars The proportion of people charging at similar charging facilities. express The predicted size of the vehicle type; then the land parcel of The charging load during a given period is expressed as follows: .

[0034] S2. Based on the seasonal variation of the installed area of ​​the distributed photovoltaic system and the solar irradiance in the region, a spatiotemporal multi-scale output model of distributed photovoltaic is constructed to predict the output of distributed photovoltaic energy and obtain the power density prediction results in the distribution network.

[0035] Preferably, the spatiotemporal multi-scale output model of distributed photovoltaic power in step S2 is expressed as follows: In the formula: express Regional distributed photovoltaic in Efforts made at all times; Indicates photoelectric conversion efficiency; This indicates the actual installed area of ​​distributed photovoltaic power generation. express Solar irradiance per unit area at any given time.

[0036] Preferably, the actual installed area of ​​distributed photovoltaic power in step S2 is... In the formula: express The available installation area for distributed photovoltaic power within the region. In the formula, express Total above-ground building area within the area express The area's floor area ratio, express Average number of floors in buildings within the area express The rate at which rooftop photovoltaic installations are possible in buildings within the region.

[0037] Preferred, This represents the floor space occupied by a single photovoltaic module. If two photovoltaic panels are considered as a single photovoltaic module and arranged in an array, then... In the formula, Indicates the minimum distance between photovoltaic arrays. This indicates the optimal tilt angle for installing photovoltaic panels. and These represent the length and width of a single photovoltaic panel, respectively. This indicates the area of ​​a single photovoltaic module; Indicates not greater than The largest integer.

[0038] S3. Based on the load density and power density prediction results, the power supply area is initially approximated and divided into input power supply area and output power supply area.

[0039] Preferably, the preliminary approximate division of the power supply area in step S3 is determined by the load and power balance results. At that time, it is an input-type power supply area, when At that time, it is an output power supply area, in which, Indicates the grid supply load. This indicates grid-connected power supply.

[0040] S4. Based on cluster analysis, determine the load center and power center in each power supply area after preliminary approximate division, and allocate the power and load in each power supply area to the nearest clustered load and power group center.

[0041] Preferably, step S4 uses an improved K-means clustering method for cluster analysis, specifically including the following steps: ① Determining the number of target clustering areas. ; In the formula: This represents the floor function. This represents the total load / power supply within the area to be divided in the distribution network. This indicates the target capacity of the zoned substations. ① Represent the power factor of the substation; ② Divide the area to be divided into A square redundant grid is used, and the center of each grid is selected as the initial load / power supply center. Choose a value greater than the required precision. A certain value; ③ Distribute all loads / power supplies to the nearest available location. ④ Eliminate the grid with the least load / power supply and allocate its load / power supply to the nearest grid. The number of redundant grids is [number missing]. Subtract 1; ⑤ Repeat step ④ until... ,get ⑥ Calculate the weighted distance from each load / power source to its center: In the formula, Indicates load / power supply To the load / power center European distance, , Indicates load / power center Distance weighting factor, In the formula, This represents the total load / power supply within the area to be divided in the distribution network. Indicates load / power center The total load / power supply obtained in the previous partitioning iteration. Indicates the weighting factor adjustment coefficient; ⑦ Distribute all loads / power sources according to the principle of minimum weighted distance to 8. Obtain the load / power center for each zone, and use the center point of each zone's load / power as the set of load / power center points for the next iteration; 9. Repeat steps 6-7 until the load / power center and zone range no longer change or the change is less than a certain precision; 10. Obtain the clustering results.

[0042] S5. Using the minimum sum of distances between load centers and power supply centers within the power supply area of ​​each substation as the objective function, and the maximum allowable load rate and power supply radius as constraints, construct an optimization model for the division of power supply areas in the distribution network.

[0043] Preferably, the objective function in step S5 is expressed as: In the formula: A set representing substation numbers; Indicates substation Load centers within service area and power center The distance between them.

[0044] Preferably, the constraint condition in step S5 is expressed as follows: In the formula: These represent the substation's load rate and maximum allowable load rate, respectively. These represent the power supply radius of the substation outgoing lines and their maximum allowable value, respectively.

[0045] S6. Solve the power supply area division optimization model of the distribution network based on the minimum weight matching method to determine the power supply range of each substation.

[0046] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention.

Claims

1. A distribution network balancing optimization method considering distributed resource consumption control, characterized in that: The method comprises the following steps: S1, based on the demand of electric vehicles in the planning area, combined with the typical charging load curve of electric vehicles and the overall scale of electric vehicles, the total charging load of electric vehicles in the planning area is predicted, and the load density prediction result in the distribution network is obtained; S2, based on the installed area of distributed photovoltaic system and the seasonal change of solar radiation value in the region, a time-space multi-scale output model of distributed photovoltaic is constructed, the output of distributed photovoltaic energy is predicted, and the power density prediction result in the distribution network is obtained; S3, based on the load density and power density prediction results, the power supply area is preliminarily approximated, including input type power supply area and output type power supply area; S4, based on the clustering analysis method, the load center and power center in each power supply area after preliminary approximation are determined, and the power and load in each power supply area are allocated to the center of each load power group after clustering; S5, taking the distance between the load center and the power center in each substation power supply area and the minimum as the objective function, and taking the maximum allowable load rate and the power supply radius as the constraint condition, a power distribution network power supply area division optimization model is constructed; S6, based on the minimum weight matching method, the power distribution network power supply area division optimization model is solved, and the power supply range of each substation is determined. 2.The distribution network balancing optimization method considering distributed resource consumption control according to claim 1, wherein: In the step S1, the Monte Carlo method is used to calculate the typical charging load curve of various electric vehicles on different charging facilities suitable for the planning area. 3.The distribution network balancing optimization method considering distributed resource consumption control according to claim 1, wherein: The total charging load of the region to be planned in step S1 is represented as: wherein: represents the total charging load of the region at time t; represents a set of electric vehicle types; represents the predicted scale of electric vehicles of type i in the region to be planned; represents the typical charging load of electric vehicles of type i, wherein: represents the typical charging load of electric vehicles of type i at time t; represents the typical charging load of electric vehicles of type i at time t; represents a set of charging facility types, represents the proportion of electric vehicles of type i that charge at charging facilities of type j; represents the proportion of electric vehicles of type i that charge at charging facilities of type j; represents the typical charging load of electric vehicles of type i at charging facilities of type j at time t; represents the typical charging load of electric vehicles of type i at charging facilities of type j at time t.​​​​​​ 4. The distribution network balancing optimization method considering distributed resource consumption control according to claim 3, characterized in that: The electric vehicle charging scale in the electric vehicle demand in the step S1 is represented as: , wherein: , respectively represent a load distribution weight and an area distribution weight, and respectively represent a conventional load value and an area of the i-th land, represent the total number of land of the functional area in the region to be planned; represent a set of electric vehicle types; represent a set of charging facility types of the i-th land; represent the number of the j-th type of vehicle charging at the k-th type of charging facility, , represent the proportion of the j-th type of vehicle charging at the k-th type of charging facility, , represent the predicted scale of the j-th type of vehicle; and the charging load of the i-th land at the h-th time period is represented as . .​​​​​​​ 5. The distribution network balancing optimization method considering distributed resource consumption control according to claim 1, characterized in that: The spatiotemporal multi-scale output model of the distributed photovoltaic in the step S2 is represented as: ; wherein: represents the output of the regional distributed photovoltaic at the time t; represents the photoelectric conversion efficiency; represents the actual installation area of the distributed photovoltaic; represents the solar radiation per unit area at the time t.

6. The distribution network balancing optimization method considering distributed resource consumption control according to claim 1, characterized in that: the actual installation area of the distributed photovoltaic in the step S2 ; wherein: represents the available installation area of the distributed photovoltaic in the region, , wherein, represents the total floor area of the buildings in the region, represents the volume rate of the region, represents the average floor of the buildings in the region, represents the installation rate of the photovoltaic on the roof of the buildings in the region; represents the floor area of a unit photovoltaic module, two photovoltaic panels are regarded as a unit photovoltaic module, and the unit photovoltaic module is arranged in an array, then , wherein, represents the minimum distance between the photovoltaic arrays, represents the optimal installation angle of the photovoltaic panels, and respectively represent the length and the width of a single photovoltaic panel; represents the area of a single photovoltaic module; represents the maximum integer not greater than .

7. The distribution network balancing optimization method considering distributed resource consumption control according to claim 1, characterized in that: The preliminary approximate division of the power supply area in step S3 is determined by the load and power supply power balance results, when it is an input type power supply area, when it is an output type power supply area, wherein, represents the grid supplied load, represents the grid supplied power supply. 8.The distribution network balancing optimization method of claim 1, wherein: The improved K-means clustering method is used in the step S4 for clustering analysis, and specifically includes the following steps: ① determining the number of clustering target supply area divisions ; , wherein: represents an integer function, represents the total load / power supply in the area to be divided, represents the target capacity of the substation, represents the power factor of the substation; ② dividing the area to be divided into square redundant grids, and selecting the center of each grid as an initial load / power supply center, taking a value greater than according to the required accuracy; ③ distributing all the load / power supplies to the initial centers; ④ removing the grid with the least load / power supply, and distributing the load / power supply to the nearest grid, and reducing the number of redundant grids by 1; ⑤ repeating step ④ until , and obtaining load / power supply centers; ⑥ calculating the weighted distance of the load / power supply to the load / power supply center: , wherein, represents the Euclidean distance of the load / power supply to the load / power supply center , , represents the distance weight factor of the load / power supply center , , wherein, represents the total load / power supply in the area to be divided, represents the total load / power supply obtained in the previous iteration of the division of the load / power supply center , represents the weight factor adjustment coefficient; ⑦ distributing all the load / power supplies to the load / power supply centers according to the principle of minimum weighted distance, obtaining the range of each division, and taking the center point position of the load / power supply of each division as the set of load / power supply center points for the next iteration; ⑧ repeating steps ⑥-⑦ until the load / power supply center and the division range no longer change or the change is less than a certain accuracy; ⑨ obtaining the clustering result.

9. The distribution network balancing optimization method considering distributed resource consumption control according to claim 1, characterized in that: The objective function in step S5 is expressed as follows: In the formula: A set representing substation numbers; Indicates substation Load centers within service area and power center The distance between them.

10. The distribution network balancing optimization method considering distributed resource consumption control according to claim 1, characterized in that: The constraint in step S5 is expressed as: wherein: respectively represent the load rate and the maximum allowed load rate of the substation; respectively represent the supply radius of the outgoing line of the substation and its maximum allowed value.

Images