Academic Journal of Science and Technology ISSN: 2771-3032 | Vol. 11, No. 1, 2024 155 Power Supply Unit Planning of Distribution Network Including Energy Storage based on N‐1 Criterion Zhan Zhang, Jin Yang* College of Electrical Engineering and automation, Henan Polytechnic University, Jiaozuo, China * Corresponding author: Jin Yang (Email: yj62082@163.com) Abstract: In order to realize effective load transfer in medium voltage distribution network when N-1 fault occurs, a method of power supply unit division is proposed. Firstly, according to the content and characteristics of grid planning of distribution network, the principle of power supply unit division is proposed. Then, based on the grid planning of distribution network, the method of power supply unit division is established. Finally, the calculation results show that the division of power supply units based on the power supply grid and energy storage planning improves the line connection rate, meets the N-1 safety check, and improves the safety and reliability of the medium-voltage distribution network. Keywords: Distribution network planning, Stored energy, Power supply unit, N-1 Security check. 1. Introduction The division of power supply units is planned on the basis of the division of power supply grids in the medium voltage distribution network and the grid-connected capacity configuration of energy storage. It is composed of multiple power supply grids, which can be transferred by the connected power supply grids in the same power supply unit in case of failure. Each power supply unit operates independently to ensure the stable and reliable operation of the power grid. It is a very important content in meshing planning of medium voltage distribution network. At present, some scholars have conducted research on the division of power supply units. Literature [1] defines the multi-level modular structure and functions at each level of the distribution network, and proposes the planning process and division method for each level. Based on the geographical boundary constraints and the quantity constraints of power supply units, literature [2] optimizes and combines the grids with high boundary coincidence degree, and proposes the evaluation index of grid boundary degree to divide power supply units. In literature [3], a mathematical model of project life cycle cost considering power failure loss factors was constructed, and a new distribution grid planning method based on capacity and power supply reliability, that is, double Q, was proposed. Literature [4] clarified the candidate channel layout in the supply area, and planned the power supply zone in the supply area based on the maximization of the "N-1 safety criterion" of the line channel and the minimization of the planned line cost. In literature [5], based on meeting the connectivity of trunk channels, and considering that load can be transferred between different lines and channels and reduce the comprehensive cost, a model was constructed, and the exhaustion method was adopted to solve it. In literature [6], the power supply units were divided based on the bipartite maximum weight algorithm[7], considering the reliability of the power supply area and the improvement of load transfer capacity, while the distance between the power supply grids was minimized and the lines of the power supply units in the power supply area were not crossed. The above research uses different methods to divide the power supply unit, which has different degrees of improvement in the performance of the power supply unit. However, in the future power grid construction, due to the change of the power grid structure, the influence of multiple factors should be considered in the division of the power supply unit, and the study on the participation of energy storage in the division of the power supply unit of the medium-voltage distribution network also plays an important role. In view of this, the power supply unit division model and method are proposed to realize the effective load transfer when N-1 fault occurs in distribution network. Firstly, based on the planning results of power supply grid and energy storage capacity of medium voltage distribution network and the characteristics of power supply area, the principle of power supply unit division is established. Then, on this basis, the power supply unit division method is established. Finally, the validity of the power supply unit division method based on power grid and energy storage planning is verified by an example analysis. 2. Principles for Dividing Power Supply Units 2.1. Power supply unit definition Power supply unit definition: consists of multiple grids with similar distances, and in the event of an N-1 fault, the interconnected grids within the power supply unit carry out belt transfer. The power supply unit operates independently, effectively reducing the impact range of faults between devices, and ensuring the safety and reliability of the medium- voltage distribution network. Power supply units are divided into power supply units on the premise of considering load transfer capacity and improving power supply reliability after power grid planning. This paper considers the supporting role of energy storage on power supply units and carries out power supply unit planning, which will be detailed in the corresponding chapters. According to the "Technical Guidelines for Distribution Network Planning and Design" issued by the National Energy Administration[8], the safety standards of medium voltage distribution network are shown in Table 1. 156 Table 1. Safety criteria of medium voltage distribution network Power supply area type Power supply safety criterion A+, A, and B class It should satisfy N-1 C class The value must be N-1 D class It satisfies N-1 E class Do not make mandatory requirements 2.2. Division principle According to the characteristics of the planned power supply area of the medium-voltage distribution network and the planning results of the meshing joint planning model of the medium-voltage distribution network based on the time series characteristics[9], each power supply grid is reasonably matched to form a power supply unit to ensure the effective load transfer in the power supply unit when N-1 failure occurs. Therefore, the principle of power supply unit division is proposed to provide a theoretical basis for the division of power supply units. According to the power supply grid planning results of distribution network and the definition of power supply unit in this paper, the power supply grid obtained after grid planning is combined into a power supply unit with uniform power supply radius, balanced load size and load balance according to geographical location. The division of power supply units should comply with the following principles. Power supply radius balance: The power supply radius of each power supply unit and the power supply grid composed of the power supply unit should be balanced to facilitate management. Load size balance: In order to improve the utilization of the equipment, the load of each power supply unit should be able to maintain a balance. Load balancing: The load ratio of each power supply grid of the power supply unit should be balanced to minimize the peak-valley difference ratio of the power supply unit, improve the load ratio, and maintain the stability of the system. 3. Power Supply Unit Division Method 3.1. Partition index After the meshing joint planning of medium voltage distribution network, in order to obtain the potential of each power supply grid, the differential matching of power supply grid is studied, so that each power supply grid can be reasonably matched to the corresponding power supply unit. In this paper, "full load difference coefficient" is proposed to quantify the matching degree between different power supply grids, and reasonable combination results are obtained through calculation, specifically:   24 2 , 1 1 7 24 i xy i X f    (1) Where: X is a positive indicator; ,i xyf indicates the load value of the power supply unit at the i time. 3.2. Partition method After obtaining the results of power supply grid division and energy storage capacity allocation through the mesh- oriented joint planning model of medium-voltage distribution network, based on the power supply grid obtained and the proposed planning index, combined with graph theory knowledge[10-11], the power supply unit planning method is proposed to meet the "N-1 safety criterion" and ensure the security and stability of the power grid. The specific process of power supply unit planning is shown in Figure 1. Start Import power grid and energy storage capacity information Establishing a Geographic Border Connection Matrix A N Y Calculate the load matrix F of each power supply unit after the power grid is sequentially switched from fault to power supply Is the number of independent grids 0? Corresponding to the boundary information in matrix A, set the load value of non adjacent grids to zero in matrix F Calculate the full load difference coefficient index X Prioritize planning power grids with high X values as one power supply unit and obtain preliminary division results Mark the grids contained in the power supply units with duplicate grids in the preliminary division results as independent grids End N Y Count the number of independent grids and find corresponding positions and values in matrices A and F Zero the corresponding positions in matrices A and F Is there a line crossing? Output power supply unit division results Figure 1. Power supply unit division flowchart The specific steps are: Step 1: Number the planned power supply grids and establish a geographic connectivity matrix  xyaA , wh ere ,x y K  . Based on graph knowledge, when two po wer supply grids are adjacent, 1xya  , x and y are the numbers of the two adjacent power supply grids and x y , otherwise they are denoted as 0. Step 2: Using the load data of each power supply gr id after planning, calculate the load matrix  xyfF aft 157 er the power supply grid switches to other power suppl y grids in case of successive faults, where x and y are the same as above. Step 3: Corresponding to the grid boundary informati on in matrix A. If there is no boundary, assign the corr esponding load value in matrix F to 0. Step 4: Calculate the degree of load pressure on the power grid after the power supply is converted. The lar ger the X value, the lower the power grid pressure. Step 5: Prioritize planning the power supply grid co mbination with the highest X value as one power suppl y unit, obtain the preliminary planning results of the po wer supply unit, determine whether there are duplicate power supply grids in each power supply unit. If there are, it cannot form a power supply unit and is tempora rily marked as an independent power supply grid. Step 6: Verify the planning results and determine if t here are any line crossings between the main power su pply lines of each power supply unit. If it exists, assig n the corresponding values in matrices A and F to 0, r eturn to step three, and if it does not exist, obtain the power supply unit planning result. Step 7: Count the number of independent power supp ly grids, search for corresponding values in matrices A and F, return to step 3, rematch, and if there are no in dependent power supply grids, end and output the plann ing result. 4. Performance Evaluation System for Distribution Networks After obtaining the results of power supply unit divis ion, this article proposes corresponding evaluation indica tors from multiple aspects to represent the degree of pe rformance improvement after distribution network planni ng. (1) Line connection rate (%): When a fault occurs in a line, the load of the faulty line can be quickly and e ffectively transferred through interconnected lines. The p roportion of connected lines in the distribution network to all lines is called the line connection rate, which is specifically: ll 1 100% L E L   (2) In the formula: llL represents the number of connected lines; L is the total number of lines. (2) N-1 pass rate (%): When an "N-1" fault occurs i n the distribution network, in order to ensure the safe a nd reliable power supply of the system, the fault load i s transferred to other lines. The proportion of lines that can achieve this transfer operation is the N-1 pass rate, specifically: hp 2 100% S E L   (3) In the formula, hpS represents the number of lines that have passed the N-1 safety criterion. (3) Average load rate of the line (%): The average v alue of the load rate at each time point during a typica l daily time period, which can reflect the load situation of the line and determine whether the load is maximize d to connect to each power grid considering the tempor al characteristics, saving the distance between outgoing lines and avoiding waste. Specifically, it is: 3 , ,lr 1 1 1 1 100% K H j h j h E f K H     (4) In the formula, , ,lrj hf represents the load rate at time h of the jth power grid. (4) Average load rate (%): This indicator is the ratio of daily average load to maximum load value, which c an reflect the utilization rate and redundancy degree of equipment in the power supply area. Specifically, it is: ,avfh 4 1 ,mf 1 100% K j j j h f E K f   (5) In the formula: ,avfhjf is the average load of the jth po wer grid; ,mfj hf is the maximum load of the jth power grid. (5) Energy saving and emission reduction rate (%): T he carbon dioxide emissions of typical daily power gen eration before and after grid planning are calculated. Th is indicator is the difference between the carbon dioxide emissions before and after planning compared to the car bon dioxide emissions before and after planning, reflecti ng the energy saving and emission reduction situation a fter planning. Specifically, it is: q h 5 q 100% e e E e    (6) In the formula: qe represents the pre planning carbon d ioxide emissions; he is the planned carbon dioxide emiss ions, and according to the statistics of China Electric P ower Union, the carbon dioxide emissions per kilowatt hour of thermal power are taken as 838 grams. 5. Example 5.1. Analysis of power supply unit division results Taking the 85 node system as an example, the voltag e level is 10kV, and the parameters are shown in refere nce[8]. The specific energy storage capacity information for supporting purposes in each power supply grid is sh own in Table 2. The power supply area is divided into 10 power sup ply grids, and a geographic adjacency matrix A is estab lished to represent the adjacency of each power supply grid. 158 Table 2. Energy storage support capacity of each power grid grid 1 2 3 4 5 6 7 8 9 10 Energy storage/MW 1.894 1.630 1.904 1.999 1.990 2.257 2.268 3.370 3.048 2.240 Establish a load matrix F for transferring power to o ther power grids in the event of a sequential failure in the power supply grid. Due to space limitations, this art icle selects matrices 6F and 18F from two of the time points. 0 1 0 1 0 0 0 1 0 0 1 0 1 0 0 0 0 1 0 0 0 1 0 0 0 0 0 1 0 1 1 0 0 0 0 1 1 1 0 0 0 0 0 0 0 1 1 0 1 0 0 0 0 1 1 0 1 0 0 0 0 0 0 1 1 1 0 1 1 0 0 1 1 1 0 0 1 0 1 1 0 0 0 0 1 0 1 1 0 1 0 0 1 0 0 0 0 1 1 0                                 A 6 0 5.38 5.5 5.39 5.53 5.52 5.34 5.56 5.18 5.33 5.38 0 5.72 5.61 5.75 5.74 5.56 5.78 5.4 5.55 5.5 5.72 0 5.73 5.87 5.86 5.68 5.9 5.52 5.67 5.39 5.61 5.73 0 5.76 5.75 5.57 5.79 5.41 5.56 5.53 5.75 5.87 5.76 0 5.89 5.71 5.93 5.55 5.7 5.52 5.74 5.86 5.75 5 F .89 0 5.7 5.92 5.54 5.69 5.34 5.56 5.68 5.57 5.71 5.7 0 5.74 5.36 5.51 5.56 5.78 5.9 5.79 5.93 5.92 5.74 0 5.58 5.73 5.18 5.4 5.52 5.41 5.55 5.54 5.36 5.58 0 5.35 5.33 5.55 5.67 5.56 5.7 5.69 5.51 5.73 5.35 0                                18 0 5.73 5.8 5.74 5.68 5.96 5.82 5.76 5.84 5.96 5.73 0 5.66 5.6 5.55 5.82 5.69 5.63 5.7 5.83 5.8 5.66 0 5.67 5.61 5.89 5.75 5.69 5.77 5.89 5.74 5.6 5.67 0 5.55 5.83 5.69 5.63 5.71 5.83 5.68 5.55 5.61 5.55 0 5.77 5.64 5.58 5.65 5.78 5.96 5.82 5.89 5.83 F 5.77 0 5.91 5.85 5.93 6.05 5.82 5.69 5.75 5.69 5.64 5.9 0 5.72 5.79 5.92 5.76 5.63 5.69 5.63 5.58 5.85 5.72 0 5.73 5.86 5.84 5.7 5.77 5.71 5.65 5.93 5.79 5.73 0 5.93 5.96 5.83 5.89 5.83 5.78 6.05 5.92 5.86 5.93 0                                Based on the established geographical adjacency matrix A and the results of the load matrix F after the power grid is sequentially converted to other power grids in case of faults, the full load difference coefficient index X is calculated to evaluate the matching degree of each power grid, and then differentiated power grid combination matching is carried out to achieve the optimal score of the power supply unit. After calculating the full load difference coefficient index X, it is converted into a percentage based matching degree matrix to represent the matching degree of the power supply units composed of each power grid. Specifically: % 0 0.2413 0.2323 0.2399 0.2357 0.2170 0.2234 0.2143 0.2158 0.2189 0.2413 0 0.2418 0.2493 0.2451 0.2264 0.2328 0.2237 0.2252 0.2283 0.2323 0.2418 0 0.2403 0.2362 0.2175 0.2239 0.2148 0.2163 0.2194 0.2399 0.2493 0.2403 0 0.2437 0.2250 0 X .2314 0.2223 0.2238 0.2269 0.2357 0.2451 0.2362 0.2437 0 0.2208 0.2272 0.2181 0.2196 0.2227 0.2170 0.2264 0.2170 0.2250 0.2208 0 0.2085 0.1994 0.2009 0.2040 0.2234 0.2328 0.2239 0.2314 0.2272 0.2085 0 0.2058 0.2073 0.2104 0.2143 0.2237 0.2148 0.2223 0.2181 0.1994 0.2058 0 0.1982 0.2013 0.2158 0.2252 0.2163 0.2238 0.2196 0.2009 0.2073 0.1982 0 0.2028 0.2189 0.2283 0.2194 0.2269 0.2227 0.2040 0.2104 0.2013 0.2028 0                                After filtering through the adjacency matrix of the po wer grid, matrix %X is used to obtain a matching degr ee map of the power supply units, as shown in Figure 2. 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 0 0 0 0 0 0 0 0 0 0 0 0 0 0.2237 0 0 0 0 0 0 0 0 0.2148 0 0.2194 0 0 0 0 0.225 0.2223 0 0 0 0 0 0 0 0.2208 0.2272 0 0.2196 0 0 0 0 0.225 0.2208 0 0.2085 0 0 0 0 0 0 0.2272 0.2085 0 0.2058 0.2073 0 0.2143 0.2237 0.2148 0.2223 0 0 0.2058 0 0.1982 0.2013 0 0 0 0 0.2196 0 0.2073 0.1982 0 0.2028 0 0 0.2194 0 0 0 0 0.2013 0.2028 0 0.2413 0.2399 0.2413 0.2418 0.2418 0.2399 0.2314 0.2314 0 0.05 0.1 0.15 0.2 Figure 2. Matching degree diagram between power supply grids From Figure 2, it can be seen that each power supply unit has a corresponding best match. When dividing power supply units, the combination with the highest X value should be prioritized to be matched within the same power supply unit. However, in the actual planning and construction process, due to geographical location or environmental constraints, it may have a certain impact on the final planning results. Therefore, it is necessary to consider the combination of the top 2 to top 3 X values between power supply grids as candidate combinations to ensure that the maximum potential for regulation between power supply grids and loads can be explored, and the goal of maximizing equipment utilization within a single power supply unit can be achieved. According to the power supply unit division method proposed in this article, the optimization division of the distribution network planning supply area was carried out, and the results of the distribution network power supply unit division are shown in Figure 3. In the figure, different grids in the power supply unit are distinguished by black line segments. 159 di st an ce /m 0 0 1000 2000 3000 4000 5000 6000 1000 2000 3000 4000 5000 transformer substation Resident ial load Administra tive load Commer cial load Health load Industria l load S distance/m [3] [2] [1] [8] [5] [7] [9] [10] [6] [4] Figure 3. Power supply unit division result diagram After planning, the indicator data of each power supp ly unit are shown in Table 3. Table 3. Power supply unit division index parameters Power supply unit Indicator value 1—4 0.2399 2—3 0.2418 5—6 0.2208 7—9 0.2073 8—10 0.2013 From Figure 3 and Table 3, it can be seen that the planned area of the medium voltage distribution networ k is divided into 10 power supply grids and 5 power s upply units. Analysis shows that after considering the te mporal characteristics of load and energy storage, the n umber of power supply grids in the planned supply are a is reduced, and the power supply units can meet the N-1 safety criterion. The equipment utilization rate is al so significantly improved, and the overall performance of the distribution network system is improved. 5.2. Analysis of performance improvement in distribution network In order to further demonstrate the efficiency improve ment of the power supply area optimization method pro posed in this article on the medium voltage distribution network, after dividing the power supply units in the pl anning area, this paper proposes to quantitatively evalua te the degree of load side efficiency improvement of th e medium voltage distribution network from five aspect s: line connection rate 1E , N-1 pass rate 2E , line average load rate 3E , average load rate 4E , and energy conserva tion and emission reduction rate 5E . The specific impro vement effects are shown in Table 4. Table 4. Comparison of indicators before and after planning index Before planning After planning Line connection rate/% 76.92 100 N-1 pass rate/% 61.54 100 Average load rate of the line/% 33.06 37.09 Average load rate/% 81.18 96.38 Energy saving and emission reduction rate/% 0 13.69 According to Table 4, the line connection rate and v arious evaluation indicators after optimizing the supply area have all been improved compared to before. The N-1 pass rate has the most significant improvement effe ct, reaching 62.50%, followed by the line connection ra te and average load rate, with improvement rates of 30. 01% and 18.72% respectively, followed by the average load rate of the line, with an improvement rate of 12.1 9%, and the energy conservation and emission reduction rate has also increased from 0 to 13.69%. It can be se en that the power supply area optimization method prop osed in this article has improved the load transfer capa city, safety, and economy of the medium voltage distrib ution network to varying degrees, reduced the redundan cy of the distribution network, improved equipment utili zation and economy, and greatly improved the overall p erformance of the distribution network. 6. Conclusion This article takes an 85 node system as an example and proposes a power supply unit division method. This method can effectively solve the problem of effective lo ad transfer when N-1 faults occur in the distribution net work, and the following conclusions are drawn: 1) The definition of power supply units has been im proved by utilizing the content and characteristics of gri d based planning for medium voltage distribution netwo rks, and the principles for dividing power supply units have been constructed. 2) Consider the evaluation indicators for power grid planning, energy storage capacity configuration, and po wer supply unit division, and propose a method for div iding power supply units. 3) Based on the evaluation indicators proposed in thi s article, the results before and after the case planning were compared, and the comparison results verified that the proposed planning method can effectively improve t he security and economy of the system. 4) The next step will be to consider more types of d istributed power sources, and further study the division of power supply units in distribution networks containin g multiple types of distributed power sources based on the model in this paper. References [1] Liu Weisheng, Liu Hong, Liu Zhiqing et al, “Modular Planning System and Method for Distribution Networks,” Journal of Power Systems and Automation.J., vol. 31, no. 11, 2019,pp. 139-144. [2] W Liu, H Liu, F Wang, et al, “Practical Automatic Planning for MV Distribution Network Considering Complementation of Load Characteristic and Power Supply Unit Partitioning,” IEEE Access.J., vol. 8, no. 4, 2020,pp. 91807-91817. [3] Hou Jia, Wu Zhili, Cui Kai, “A New Grid Planning Method for Distribution Networks Based on Double Q,” Guangdong Electric Power.J., vol. 32, no. 04, 2019,pp. 17-25. [4] Ren Hongyu, Wang Zhuding, Zhang Chao, et al, “Optimization Model and Method for Grid Planning of High Voltage Distribution Networks,” Power System Automation.J., vol. 43, no. 14, 2019,pp. 151-158. [5] Zhang Man, Wang Zhuding, Li Qiang, et al, “Optimization model and method for power supply zoning in medium voltage target network planning,” Power System Automation.J., vol. 43, no. 16, 2019,pp. 125-135. 160 [6] Tang Bo, Yang Peng, Yu Guangzheng, et al, “Optimization method for medium voltage distribution network supply area based on load peak valley coupling characteristics,” Chinese Journal of Electrical Engineering.J., vol. 42, no. 19, 2022,pp. 7051-7063. [7] Wang Haiying, Huang Qiang, Li Chuantao, et al, “Graph theory algorithms and their MATLAB implementation,” Beijing: Beihang University Press.M.,2010,pp. 101-105. [8] Technical Guidelines for Distribution Network Planning and Design: DL/T 5729-2016, National Energy Administration.S. 2016. [9] Zhang Z, Yang J, “Grid Planning Method of Medium Voltage Distribution Network Including Distributed Power Sources Based on Timing Characteristics,” IJEPES.J., vol. 2, no. 1, 2024,pp. 11-24. [10] Yao Yongfeng, Wang Qizhe, Wang Huiping, et al, “Fault section localization method for low-voltage distribution networks based on adjacency matrix,” China Electric Power.J., vol. 54, no. 11, 2021,pp. 91-96114. [11] Xu Jingyou, Ding Jianyong, Yang Dongjun, et al, “A reliability evaluation method for distribution networks based on graph theory and network equivalence,” Journal of Electrical Engineering.J., vol. 16, no. 03, 2021,pp. 92-98.