Optimization method of power supply radius length of distribution network based on CIM

Abstract

A CIM-based distribution network power supply radius length optimization method uses CIM model data and the breadth-first search (BFS) algorithm to quickly calculate the power supply radius of all distribution transformers on the feeder. By screening, a local optimal method suitable for engineering applications is obtained, and a reasonable solution to the problem of excessive power supply radius is quickly formed.

Description

Technical Field

[0001] The present invention relates to a technology in the field of distribution network control, specifically a method for optimizing the power supply radius length of a distribution network based on CIM (Common Information Model). Background Art

[0002] As a critical link in power grid operation, the difficulty in solving grid problems lies in obtaining and analyzing topology. Currently, China Southern Power Grid uses the CIM model as the data structure for storing distribution network topology information. This model has a large amount of data, is complex, and has a difficult data structure to analyze and process. The problem of excessively long power supply radius is a common grid problem in distribution networks. Excessive power supply radius can lead to degraded power quality at the end of the line, such as low voltage. Solving the problem of excessive power supply radius for individual distribution transformers can rely on manual methods, such as treating specific cases in a specific manner to find the optimal solution. However, a universal method for large-scale and diverse applications has yet to be proposed.

[0003] The issue of an excessively long power supply electrical radius refers to the electrical distance between medium-voltage users and the feeder outlet of a substation. Excessive distance can cause power quality issues such as low voltage, thereby reducing the overall reliability of the distribution network. According to the China Southern Power Grid Co., Ltd. corporate standard, "Technical Guidelines for 110 kV and Below Distribution Network Planning," the power supply distance of medium-voltage distribution lines must meet terminal voltage quality requirements. The recommended range for the power supply electrical radius is shown in Table 1.

[0004] Table 1 Power supply area division table

[0005]

[0006] Note 1: σ is the load density of the power supply area in the planned horizontal year (MW / km2).

[0007] Note 2: The power supply area should not be less than 5km2.

[0008] Note 3: When calculating load density, the dedicated line load of 110kV and above voltage levels, as well as the ineffective power supply areas such as mountains, Gobi, deserts, waters, and forests should be deducted.

[0009] Note 4: Regional levels are divided into four tiers based on administrative level, city importance, economic status, and load density. The 15 cities are: Guangzhou, Shenzhen, Foshan, Dongguan, Zhuhai, Nanning, Guilin, Liuzhou, Kunming, Qujing, Honghe, Guiyang, Zunyi, Haikou, and Sanya. Guangzhou and Shenzhen are international metropolises.

[0010] Table 2 Recommended power supply electrical radius

[0011]

[0012] According to different power supply areas and distribution networks, the threshold value for excessive power supply radius is set.

[0013] The solution to the problem of an excessively long power supply radius is to reroute the distribution transformer cluster within the station or transfer adjacent lines. This solution requires further determination of the transfer method (within the station or between stations) and the transfer point. The problem of an excessively long power supply radius requires determination of the transfer method (within the station or between stations), the starting point, and the end point of the transfer point. Summary of the Invention

[0014] In response to the above-mentioned shortcomings of the existing technology, the present invention proposes a CIM-based distribution network power supply radius length optimization method. By utilizing CIM model data and the breadth-first search (BFS) algorithm, a rapid calculation of the power supply radius of all distribution transformers on the feeder is formed. By screening, a local optimal method suitable for engineering applications is obtained, and a reasonable solution to the problem of excessive power supply radius is quickly formed.

[0015] The present invention is achieved through the following technical solutions:

[0016] The present invention relates to a method for optimizing the power supply radius length of a distribution network based on CIM, comprising:

[0017] Step 1) Collecting distribution network CIM data: calling the CIM file in the power grid system and parsing it into a database file using code;

[0018] Step 2) using the common information model (CIM) of the power system to analyze the distribution network topology link relationship and envelope data calculation;

[0019] Step 3) Find a suitable transfer point within the station, i.e., the starting point of the transfer line; find the adjacent line based on the geographical range of the transfer point within the station; then find a suitable transfer point outside the station on the adjacent line, i.e., the end point of the transfer line, thereby calculating the possible power supply radius transfer method on a global scale. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a schematic diagram of the topological structure of the CIM model of the distribution network;

[0021] Figure 2 This is a schematic diagram of the optional range of transfer points within the station;

[0022] Figure 3 This is a schematic diagram of the breadth-first search algorithm;

[0023] Figure 4 Schematic diagram of the embodiment effect. DETAILED DESCRIPTION

[0024] like Figure 1 As shown, this embodiment relates to a method for optimizing the power supply radius length of a distribution network based on CIM. Taking a large city in southern China as an example, cities of this size generally have thousands of feeders and tens of thousands of 10kV medium-voltage distribution transformers. Each feeder has approximately 50-500 overhead or cable lines, and the longitude and latitude coordinates of these lines on each feeder range from 500 to tens of thousands. The method specifically includes:

[0025] Step 1) Collect distribution network CIM data: call the CIM file in the power grid system and use the code to parse it into a database file.

[0026] Step 2) Use the common information model (CIM) of the power system to analyze the topological link relationship of the distribution network and calculate the envelope data: Utilize the characteristics of the BFS algorithm to calculate the power supply radius length of all distribution transformers globally.

[0027] like Figure 1 As shown, the distribution network topology link relationship refers to the logical definition of the topological relationship of distribution network equipment, including: Terminal (endpoint class), Connectivity node (connection node class), Topological node (topological node class), Topological island (topological island), and Conducting equipment (conducting equipment class). A Terminalall belongs to a Conducting equipment, but a Conducting Equipment may have any number of Terminals, that is, a Terminal provides one or more external connections for the Conducting Equipment. Each Terminalall can be connected to zero or one Connectivity node. Connectivity node / is the point where the endpoints of the conductive equipment are connected together through zero impedance. Terminalali are connected through Connectivity node, which describes the switch-node model of the actual power system. A Connectivity node can contain any number of Terminals and can also be a member of a Topological nodef. A Topological node is a dynamic node generated by integrating Connectivity nodes. It actually corresponds to the bus in the bus-branch model. A Topological island corresponds to the actual electrical island in the power system and is composed of a group of Topological nodesl with electrical connection characteristics.

[0028] So, in summary, in the original data model, two connected devices need to perform four traversal searches to find each other, which is extremely time-consuming when the data scale is large. This patent first simplifies this connection method and eliminates some redundant data.

[0029] The analysis mentioned above refers to the following steps: first, the CIM model is simplified from two aspects: data scale and link mode. Specifically, in terms of data scale, some unconnected topologies and some device information irrelevant to this problem are removed. Taking this embodiment as an example, the original "Lingshan F24" has 12,000*7 data items, but after simplification, there are only 300*14 data items. The link mode uses the BFS algorithm to simplify the original four-segment link relationship into one segment, and removes the Terminal and Connectivity nodes. This can increase the subsequent computing efficiency by four times, facilitating large-scale applications.

[0030] like Figure 3 As shown, in a multi-node undirected graph, node A is the starting node. The breadth-first search algorithm based on python code will create a state queue BFS. The starting state in BFS is node A. The characteristic of the queue data structure is "first in, first out", that is, the element that enters the queue first will be processed first. The breadth-first search algorithm will process each node stored in the state queue BFS. The processing order follows the "first in, first out" principle. Each processing will cause the processed node to leave the queue, and the "child node" of the processed node will be added to the end of the queue. Figure 3 For example, initially, the BFS will store the information of node A. The breadth-first search algorithm will first process node A, put its two "child nodes" B and C into the queue BFS, and then delete node A from BFS. Then, the queue now contains two new nodes B and C, so the breadth-first search algorithm will not stop, but will continue to process node B, put all the child nodes of node B, namely D and E, into the queue BFS, and then delete node B. Repeating this process will eventually traverse all the nodes in the entire graph. Figure 3 For example, the final output result is ABCDEFGHIJKLMX, which means the entire graph is traversed layer by layer.

[0031] In the original CIM model topology structure, the links between various devices are through Terminal and CN nodes, and the "sub-nodes" associated with each node can be accurately found. In the breadth-first search algorithm, the output method also follows the logical structure of "parent node" and "sub-node". Therefore, when using the breadth-first search algorithm to traverse the data parsed from the original CIM model, the output results can include the topology structure information of the entire line. Through the data reconstruction of the BFS algorithm, the four-layer link method of the original CIM model is changed to one layer, greatly improving the operation efficiency and providing the possibility for large-scale engineering applications.

[0032] Step 3) Calculate the method for shortening the power supply radius: By finding a suitable transfer point within the station, that is, the starting point of the transfer line; according to the geographical location range of the transfer point within the station, find the adjacent line, and then find a suitable transfer point outside the station on the adjacent line, that is, the end point of the transfer line, so as to calculate the possible power supply radius transfer method within the global scope, specifically including:

[0033] 3.1) Find the most suitable transfer point within the station: To reduce costs, the transfer point is generally selected near the sectionalizing switch. As Figure 2 shown, according to the feeder topology structure, the standard transformers are all the distribution transformers with the problem of excessive power supply radius on this feeder. Node A is the outlet of the line. Node B is the sectionalizing point where the power supply radius threshold is located. The meaning of point B as the threshold point is that the electrical length of all distribution transformers from point B is less than the power supply radius threshold and the sectionalizing point closest to the outlet A point. That is, there is a distribution transformer, and the electrical length of any upper-level line from B is greater than the set threshold. Therefore, point B is the upper limit of the transfer point, and all possible transfer points are at the back end of B. Point C is the best transfer point within the station. Point C has two characteristics. One is that it is on the lower-level line of point B, and the other is that point C is the first common upper-level line of all distribution transformers with excessive power supply radius. Obviously, point C is the best transfer point for this line. The reasons are as follows: Taking point D as a comparison, assuming the other end of the transfer is J, then no matter where J is, there is always CJ < CD + DJ. The final transfer point will be generated between point B and point C, and due to the superiority of point C, it will be as close as possible to point C according to the actual transfer position and transfer line length requirements.

[0034] 3.2) Find a suitable adjacent transfer point within the global scope: When the global data scale is large, finding the shortest distance among a large number of coordinate values is a problem with a time complexity of O(mn), where m is the possible transfer points between point C and point B, and n is the transfer points within the global scope, and this value is greater than 10 6 ,

[0034] , and this is only the amount of calculation for one feeder. It can be foreseen that if a direct search method is used, the calculation time will be greatly extended. Therefore, this embodiment uses a step-by-step calculation method to simplify the previous process and obtain a local optimal result. Specifically: Since the actual topological structure of the feeder has different shapes, most of them are irregular tree-like, so this embodiment first obtains the rectangular envelope coordinate values ​​of each feeder as an intermediate result and stores them in advance, and then calculates the possible transfer points and the positions of each envelope in the station, and selects the K feeders closest to the transfer points in the station, reducing the thousands of feeders globally to K, and then traverses the distances between the K feeders and the transfer points in the station and obtains: 1) the length of the new transfer line to be added, 2) according to the position of the off-station transfer point in the new line, the length of the power supply radius after the transfer is obtained. Thus, the final optimal transfer solution is obtained to solve the problem of excessive power supply radius.

[0035] This embodiment is further deployed in Guangzhou to calculate and solve the problem of global power supply radius, such as Figure 4 As shown, first identify the distribution transformer information that the power supply radius of the station is too long, and the calculation results are shown in Table 3

[0036] Table 3 List of distribution transformers with excessive power supply radius

[0037] Transformer name Id Power supply electrical radius 'Tianxx.10kV.Winding', '65xxxxxxxxxxx313', '8012.2597' 'Nan Er xx.10kV.Winding', '65xxxxxxxxxxx813', '8073.2597' 'Nan Ersan xx.10kV. Winding '65xxxxxxxxxxx913' '8867.2597' 'Six hectares xx.10kV.winding', '65xxxxxxxxxxx913', '10795.2597' 'Nan Ercun xx.10kV.Winding', '65xxxxxxxxxxx813', '11619.2597

[0038] The station belongs to the B-type feeder, and the power supply radius threshold is 6km. After identification, there are 5 distribution transformers whose power supply radius lengths do not meet the requirements, such as Figure 4 Based on the distribution of the five distribution transformers, calculations show that the power supply radius threshold is the location marked by the yellow box, and the optimal intra-station transfer point is the location marked by the green box.

[0039] Table 4 Optimization results of the problem of too long power supply radius

[0040] Problem feeder name 110kV Lingshan Substation_Lingshan F24 South Second Line Problem feeder ID 65xxxxxxxxxx801 Latitude and longitude of the link point within the site (113.4549622，22.7682729) Transfer to adjacent line name 110kV Hengli Substation_Hengli F15 Xinxing Line Transfer adjacent line ID 65xxxxxxxxxx101 Latitude and longitude of external link points (113.460493，22.7475155) Length of new transfer line / m 2379.4312 Shorten distance / m 6212.7623

[0041] It can be seen from the calculation results of Lingshan F24 that the power supply radius of the five distribution transformers can be effectively shortened to within 6km.

[0042] This example was tested on the 7474 feeders in a Python 3.7.5 environment, using only a single-core processor and 8GB of memory. A total of 7046 feeders were found to be valid (invalid feeders, for example, due to factors such as the lack of a distribution transformer). The program ultimately ran for 5349 seconds (approximately one and a half hours). This approach is suitable for rapid, large-scale, real-time monitoring of the problem of excessively long power supply radius in the global distribution network of large cities, and for providing solutions.

[0043] Compared with the prior art, the present invention achieves a universal and program-friendly reduction in the power supply radius of the distribution transformer through a fast calculation method (CIM model simplification and feeder envelope longitude and latitude) under a large number of feeders.

[0044] The above-mentioned specific implementation can be partially adjusted in different ways by those skilled in the art without departing from the principles and purpose of the present invention. The scope of protection of the present invention shall be based on the claims and shall not be limited by the above-mentioned specific implementation. All implementation schemes within its scope shall be subject to the constraints of the present invention.

Claims

1. A method for optimizing the power supply radius length of a distribution network based on CIM, characterized in that: include: Step 1) Collecting the Common Information Model (CIM) data of the distribution network: calling the CIM file in the power grid system and parsing it into a database file using code; Step 2) using the common information model (CIM) of the power system to analyze the topological link relationship of the distribution network and calculate the envelope data; Step 3) by finding a suitable transfer point within the station, i.e. the starting point of the transfer line; Based on the geographical location of the on-site transfer point, the adjacent line is found. Then, on the adjacent line, a suitable off-site transfer point (i.e., the end point of the transfer line) is found. This allows the candidate power supply radius transfer methods to be calculated globally, specifically including: 3.1) Find the most suitable transfer point within the station; 3.2) A step-by-step calculation method is used globally to find a suitable nearby off-site transfer point. Specifically, the rectangular envelope coordinates of each feeder are first obtained and stored as intermediate results. The positions of the candidate on-site transfer points and each envelope are then calculated. The K feeders closest to the on-site transfer points are screened out, reducing the global number of thousands of feeders to K. The distances between these K feeders and the on-site transfer points are then traversed to determine: 1) the length of the newly added transfer line; 2) the length of the power supply radius after the transfer, based on the position of the off-site transfer point on the new line. This ultimately leads to the optimal transfer solution to address the problem of an excessively long power supply radius.

Images