Layered planning method of three-terminal low-voltage DC remote power supply system for power distribution network optimization

Abstract

The application provides a three-terminal low-voltage DC remote supply system hierarchical planning method for power distribution network optimization, comprising: collecting low-voltage power distribution network planning and operation basic element data; based on the low-voltage power distribution network planning and operation basic element data, constructing a three-terminal low-voltage DC remote supply system upper layer planning model with a minimum system operation cost as an objective function, the system operation cost being a comprehensive investment, maintenance cost and lower layer feedback of the low-voltage DC remote supply system; constructing a three-terminal low-voltage DC remote supply system lower layer optimization model with a minimum power distribution network power purchase cost, network loss cost and voltage penalty cost as an objective function; and performing cyclic iteration solution on the three-terminal low-voltage DC remote supply system upper layer planning model and the three-terminal low-voltage DC remote supply system lower layer optimization model to determine a three-terminal low-voltage DC remote supply system hierarchical planning scheme. The method provided in the application effectively solves the user voltage out-of-limit problem by coordinating and optimizing the three-terminal low-voltage DC remote supply system, reduces the investment and operation cost, and improves the economic benefit.

Description

Technical Field

[0001] This application belongs to the field of power distribution network technology, and in particular relates to a hierarchical planning method for a three-terminal low-voltage DC remote power supply system for distribution network optimization. Background Technology

[0002] To reduce greenhouse gas emissions, the penetration rate of distributed energy, represented by distributed photovoltaics, in power distribution networks is continuously increasing. However, power distribution networks, especially rural power grids with long power supply radii and end-point lines, face increasingly prominent voltage quality problems. During peak load periods, low voltage is prone to occur at the end of the line, seriously affecting the safe operation of users' electrical equipment. This contradiction severely restricts the carrying capacity of the power distribution network for distributed energy, leading to increased network losses and decreased power quality. The essence of voltage quality problems is the power imbalance in the distribution network in the spatiotemporal dimensions. Traditional measures such as capacitor reactive power compensation and line upgrades have limited regulation capabilities and poor economic efficiency. In contrast, low-voltage DC remote power supply systems utilize the three-phase active / reactive power decoupling and flexible control capabilities of voltage source converters (VSCs) and the advantage of larger DC transmission capacity. Through series and parallel connection methods, high / low voltage regulation can be achieved. However, existing research has not considered using low-voltage DC remote power supply systems to solve voltage quality problems; and it focuses more on independent planning or simple sequential planning, failing to establish a direct link between voltage quality and investment and operational economy, resulting in planning schemes that either over-invest or have insufficient regulation effect; in addition, the operational constraints of DC remote power supply systems are not adequately considered in the modeling, especially the coordinated operation strategy, which affects the practicality and economy of the planning schemes. Summary of the Invention

[0003] This application provides a hierarchical planning method for a three-terminal low-voltage DC remote power supply system for distribution network optimization. To solve the above-mentioned technical problems, this application adopts the following technical methods: Firstly, this application provides a hierarchical planning method for three-terminal low-voltage DC remote power supply systems for distribution network optimization, including: Collect basic data on low-voltage distribution network planning and operation; Based on the basic data of low-voltage distribution network planning and operation, an upper-level planning model for low-voltage DC remote power supply system is constructed with the objective function of minimizing the comprehensive investment, maintenance cost and lower-level feedback system operation cost of the low-voltage DC remote power supply system. Construct a lower-level optimization model for a low-voltage DC remote power supply system with the objective function of minimizing the power purchase cost, network loss cost, and voltage penalty cost of the distribution network; The upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system are solved iteratively to determine the hierarchical planning scheme of the three-terminal low-voltage DC remote power supply system.

[0004] Optionally, the basic data for low-voltage distribution network planning and operation includes the grid structure and load information of the low-voltage distribution network system. The grid structure of the low-voltage distribution network system includes the impedance information of each branch. The load information includes the active and reactive loads of each node in the low-voltage distribution network.

[0005] Optionally, the constraints of the upper-level planning model of the low-voltage DC remote power supply system include the following: VSC control mode constraints, access location and capacity constraints, and line length constraints.

[0006] Optionally, the constraints of the lower-level optimization model of the low-voltage DC remote power supply system include the following: Branch power flow constraints, node voltage constraints, and power constraints; wherein, the branch power flow constraints include AC power flow constraints and DC power flow constraints, and the power constraints include DC power balance constraints and AC power balance constraints.

[0007] Optionally, the step of iteratively solving the upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system to determine the hierarchical planning scheme of the three-terminal low-voltage DC remote power supply system includes: A hybrid optimization algorithm combining genetic algorithm and second-order cone programming is used to iteratively solve the upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system to determine the hierarchical planning scheme of the three-terminal low-voltage DC remote power supply system.

[0008] Optionally, the method employs a hybrid optimization algorithm combining genetic algorithm and second-order cone programming to iteratively solve the upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system, thereby determining a hierarchical planning scheme for the three-terminal low-voltage DC remote power supply system, including: A genetic algorithm is used to solve the upper-level planning model of the low-voltage DC remote power supply system and output the optimal addressing and capacity setting scheme. By employing variable substitution and convex relaxation, a second-order cone relaxation transformation is performed on the lower-level optimization model of the low-voltage DC remote power supply system to obtain a second-order cone lower-level programming model in convex optimization form. Substitute the optimal addressing and occupancy scheme into the second-order cone lower-level planning model for solution, and output the optimal scheduling strategy; The upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system are iteratively optimized until the iterative optimization termination condition is met, and the hierarchical planning scheme of the three-terminal low-voltage DC remote power supply system is output.

[0009] Optionally, the iterative process of the genetic algorithm includes the design of decision variables and hybrid encoding, the construction and calculation of the fitness function, selection operation, crossover operation, and mutation operation; wherein, the selection operation adopts roulette wheel selection and elite retention selection; the crossover operation adopts categorized crossover; the binary part of the mutation operation adopts bit-flip mutation; the calculation result of the fitness function directly maps the optimization objective achievement degree of the upper-level planning model of the low-voltage DC remote power supply system.

[0010] Secondly, this application also provides a computer system, comprising: Memory is used to store instructions that can be executed by the processor; A processor for executing the instructions to implement the method as described in the first aspect.

[0011] Thirdly, this application also provides a computer-readable medium storing computer program code that, when executed by a processor, implements the method described in the first aspect.

[0012] This application has the following beneficial effects: The method proposed in this application effectively solves the problem of user voltage exceeding limits by coordinating and optimizing a three-port low-voltage DC remote power supply system, while reducing investment and operating costs and improving economic efficiency. Attached Figure Description

[0013] Figure 1 A flowchart illustrating the hierarchical planning method for a three-terminal low-voltage DC remote power supply system for distribution network optimization provided in this application embodiment; Figure 2 This is a network topology diagram of a low-voltage distribution network provided in an embodiment of this application; Figure 3 A schematic diagram of a three-terminal parallel low-voltage DC remote power supply system provided in the embodiments of this application; Figure 4 The convergence curve of the genetic algorithm for hierarchical planning is provided in the embodiments of this application; Figure 5 This is a diagram showing the distribution of voltage box lines under different schemes provided in the embodiments of this application. Detailed Implementation

[0014] To facilitate understanding by those skilled in the art, the present application will be further described below in conjunction with embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present application.

[0015] To solve the above technical problems, such as Figure 1 As shown, this application proposes a hierarchical planning method for three-terminal low-voltage DC remote power supply systems for distribution network optimization, including: S101: Collect basic data on low-voltage distribution network planning and operation; Basic data for low-voltage distribution network planning and operation include the network structure and load information of the low-voltage distribution network system. The network structure of the low-voltage distribution network system includes the impedance information of each branch, such as the resistance of AC branches. Reactance DC branch resistance Load information includes the active load of each node in the low-voltage distribution network. P load,ac reactive load Q load,ac .

[0016] Since the grid structure and load information of the aforementioned low-voltage distribution network system will affect the cost of the low-voltage distribution network, the corresponding data is obtained here in order to build the corresponding model in the future.

[0017] S102: Based on the basic data of low-voltage distribution network planning and operation, construct an upper-level planning model for the low-voltage DC remote power supply system with the objective function of minimizing the comprehensive investment, maintenance cost and lower-level feedback system operation cost of the low-voltage DC remote power supply system. Based on the aforementioned basic data on low-voltage distribution network planning and operation, a system is constructed with the objective function of minimizing the comprehensive investment, maintenance costs, and operating costs of the low-voltage DC remote power supply system and the lower-level feedback system. The upper-level planning model of the low-voltage DC remote power supply system is shown below: (1) (2) (3) In the formula: This refers to the set of nodes included in all low-voltage DC remote power supply systems. C RPS , C OM,RPS The annual equivalent investment cost and annual maintenance cost of the low-voltage DC remote power supply system; r The discount rate for the equipment; y RPS The service life of low-voltage DC remote power supply equipment; c p,host and c p,slave Investment cost per unit capacity of the master and slave units in a parallel low-voltage DC remote power supply system; c line The investment cost per unit length of low-voltage DC long-distance power supply lines; S i,host , S j,slave and Lij These refer to the installation capacity of the main unit and slave unit of the low-voltage DC remote power supply system, and the length of the DC remote power supply line. c m,host , c m,slave These are the annual maintenance costs per unit capacity for the main unit and slave unit of the low-voltage DC remote power supply system, respectively.

[0018] The constraints of the above-mentioned upper-level planning model for low-voltage DC remote power supply system include the following: 1) VSC control mode constraints: A three-port parallel low-voltage DC remote power supply system requires a VSC as the master unit, which operates in... U dc - Q In the control mode, the main function is to stabilize the DC voltage. The other two VSCs can act as slaves to handle power regulation, i.e., PQ mode. Therefore, control mode constraints need to be established, and the model is as follows: Define the pattern variable as m i ∈{0,1}, m i =1 is U dc - Q In control mode, the corresponding VSC is the host. m i =0 is PQ Control mode, when VSC is in U dc - Q The control mode can maintain the DC voltage while being in a state of... PQ In control mode, active and reactive power can be flexibly adjusted, subject only to power balance and capacity constraints. Because each VSC in the low-voltage DC remote power supply system is connected via a DC line, only one VSC is needed as the main unit. U dc - Q The control mode maintains the DC voltage. Therefore, the control mode constraint can be modeled as follows: (4) In the formula: N is the number of distribution network nodes.

[0019] 2) Access location and capacity constraints When selecting the VSC access locations in a three-port parallel low-voltage DC remote power supply system in a distribution network, it is necessary to ensure that the number of master units is 1 and the number of slave units is 2, and that the master unit access location and the two slave unit access locations are all different. Therefore, its installation location constraint model can be established as follows: (5) In the formula: y iIt is a 0-1 variable, and its value of 0 represents a node. i Without VSC installed, a value of 1 indicates a node. i Install VSC; the first term in the formula indicates that the total number of VSC installations is 3; the second and third terms in the formula indicate when... y i =1 and m i When =1, the node connects to the host; when... y i =1 and m i When = 0, it is a slave device; when both are 0, it is not connected to VSC.

[0020] The master and slave units of the low-voltage DC remote power supply system need to meet certain capacity requirements. Considering the access location, the capacity constraint model is as follows: (6) In the formula: , These are the minimum installation capacities for the master and slave devices, respectively. , These are the maximum installed capacity for the master and slave devices, respectively.

[0021] 3) Line length constraints Furthermore, the DC line length of the low-voltage DC remote power supply system is related to the location of the VSC access nodes at both ends. It can be derived by accumulating the AC branch lengths between distribution network nodes, and considering losses due to line detours and bends, a line laying correction factor is set. Therefore, assuming the main unit is denoted as VSC1, and slave units 1 and 2 are denoted as VSC2 and VSC3 respectively, the DC line length can be calculated as follows: (7) In the formula: , They are nodes i and j、i and k The set of communication branches between them; l m branch road m Length; , These are the correction factors for laying DC lines between the main unit and slave units 1 and 2, respectively.

[0022] Furthermore, once the installation location is determined, the length of the DC line between the master and slave units must be within a certain range, and its constraint model is as follows: (8) In the formula: L max This represents the maximum length of the line.

[0023] Therefore, the length of the DC line is affected by the connection location of the low-voltage DC remote power supply (VSC). When the connection location (yi) changes, the length of the DC line (Lij) will also change, thus affecting the DC resistance. This leads to network loss. Things have changed.

[0024] S103: Construct a lower-level optimization model for a low-voltage DC remote power supply system with the objective function of minimizing the power purchase cost, network loss cost, and voltage penalty cost of the distribution network; The lower-level optimization model is a sub-problem of distribution network operation optimization aimed at improving the voltage quality of the distribution network. Its objective function F is to minimize the distribution network's electricity purchase cost, network loss cost, and voltage penalty cost. down The lower-level optimization model of the low-voltage DC remote power supply system is shown below: (9) (10) (11) (12) In the formula: C G The cost of purchasing electricity from the upper-level power grid for the generator; C L Costs related to network losses; C V This is a penalty item for voltage exceeding the limit; The price for purchasing electricity from the higher-level power grid; This represents the active power injected from the upstream power grid into the distribution network during time period t on the z-th typical day. , These are the line losses for the AC branch and the DC branch, respectively; V min V max- These are the upper and lower limits of the node voltage, respectively; , These are the costs of AC and DC branch line losses, respectively. , These represent the penalty costs for voltages exceeding the upper and lower limits, respectively. Ω G , Ω load , Ω ac and Ω dc These are collections of generators, loads, AC lines, and DC lines, respectively. Ω d and Ω T These are sets of typical daily scenarios and simulated runtime segments, respectively. Let be the voltage amplitude at node i during time period t on typical day z. This represents the active power of the load connected to AC node i during time period t on typical day z.

[0025] The model constraints for the lower-level optimization mentioned above include the following: 1) Branch flow constraints a) Constraints of the trend of communication (13) (14) (15) (16) In the formula: α(i) represents the node i Let β be the set of the end nodes of the branch with the first node; β(i) is the set of the branch nodes with the first node. i FVR represents the set of branch start nodes of the last node; FVR∈{host,slave}, which are the low-voltage DC remote power supply host and slave respectively; and These are the active and reactive power of AC branch ji, respectively; and These represent the active and reactive power injected into AC node i, respectively; The magnitude of the current flowing through AC branch ij; The voltage amplitude at node i during time period t; and These are the resistance and reactance of AC branch ij, respectively; , These are the active and reactive power injected into the substation connected to AC node i, respectively. , These refer to the active and reactive power injected by the master or slave unit of the low-voltage DC remote power supply system connected to AC node i, respectively. , These represent the active and reactive power of the load connected to AC node i, respectively.

[0026] b) DC power flow constraints (17) (18) (19) (20) In the formula: The active power of the DC branch ji; The active power injected into DC node i; The magnitude of the current flowing through the DC branch ij; The voltage amplitude at DC node i during time period t; Let be the resistance of the DC branch ij; The active power injected by the master or slave of the low-voltage DC remote power supply system connected to DC node i; This represents the load power connected to DC node i.

[0027] 2) Node voltage constraints: The AC node voltage must meet the following constraints: (twenty one) In the formula: Let be the voltage amplitude of AC node i during time period t on typical day z; and These are the upper and lower limits of the AC node voltage amplitude, respectively.

[0028] Similarly, the constraint model for DC node voltage is: (twenty two) In the formula: Let be the voltage amplitude of DC node i during time period t on typical day z; and These are the upper and lower limits of the DC node voltage amplitude, respectively.

[0029] 3) Power constraints For the DC line junction side, a DC power balance constraint can be established as follows: (twenty three) In the formula: , These represent the DC-side power of the master and slave units of the low-voltage DC remote power supply system during the t-hour period on a typical day (z). This refers to DC line losses.

[0030] For the AC line junction side, AC power balance constraints can be established as follows: (twenty four) (25) In the formula: , These represent the AC power of the master and slave units of the low-voltage DC remote power supply system during the t-hour period on a typical day (z). , These are the active power losses of the main unit and slave unit of the low-voltage DC remote power supply system, respectively. , These are the power loss coefficients of the master and slave units, respectively; , These refer to the capacities of the main unit and slave unit of the low-voltage DC remote power supply system.

[0031] S104: Iteratively solve the upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system to determine the hierarchical planning scheme of the three-terminal low-voltage DC remote power supply system.

[0032] This step mainly employs a hybrid optimization algorithm combining genetic algorithm and second-order cone programming to iteratively solve the upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system, thereby determining the hierarchical planning scheme for the three-terminal low-voltage DC remote power supply system. The specific process is as follows: A genetic algorithm is used to solve the upper-level planning model of the low-voltage DC remote power supply system, thereby outputting the optimal addressing and capacity determination scheme. The calculation process is as follows: First, we need to design the decision variables and hybrid coding: Decision variables include location variables y i , volumetric variables S VSC,i and line length variable L ; y i ∈{0,1}, is a discrete variable, and its value of 1 indicates that at node 0, 1 ... i Install it only if specified, otherwise do not install it, and all y i The total number is 3; the capacitive variable is a continuous variable, and when not connected to VSC, that is... y i When =0, S VSC,i =0;

[0033] (26) In the formula N The number is a binary code representing the addressing variable, i.e., the installation location of the master and slave devices, with a total of 3. The middle three bits are integer codes representing the capacity variable; the last two bits are real number codes representing the line length. m host For host identifier bits, m host ∈{1,…, N}(Directly encode the selected host node number, satisfying) m host ∈{ i | y i =1}). Furthermore, encoding verification is required; that is, when the address node changes, it needs to be recalculated in real time.

[0034] The fitness function is then constructed and calculated. The calculation result of the fitness function generally directly maps the degree of achievement of the optimization objective of the upper-level planning model of the low-voltage DC remote power supply system, as follows: (27) In the formula, This is the fitness function value for the k-th iteration. This represents the value of the upper-level objective function in the k-th iteration.

[0035] Finally, genetic operations are performed, including selection, crossover, and mutation. The selection operation employs a roulette wheel selection strategy and an elite retention strategy, preserving the top 10% of optimal individuals to avoid losing high-quality solutions. The roulette wheel selection probability is inversely proportional to applicability, ensuring that low-cost solutions are selected first.

[0036] (28) In the formula: M is the population size, Fitness max This represents the maximum fitness of the current generation.

[0037] The crossover operation employs a type-based crossover method. For the binary portion (address variables), single-point crossover is performed, with the crossover point randomly selected. After crossover, the total number of address variables is verified to be 3. (29) In the formula: c Intersection point; p , q For the parent generation; k , l For offspring individuals.

[0038] For the real part (capacity and line length), arithmetic crossover is used to preserve capacity boundary constraints: (30) In the formula: α∈[0,1] is the random weight.

[0039] In the mutation operation, the binary part uses bit-flipping mutation, while the integer encoded segment uses uniform mutation. The mutation logic for positional variables is: randomly flip a bit... i If the sum of the totals is not 3, then flip one compensation node.

[0040] For each lower-level individual, since the original distribution network operation optimization equations are nonlinear and nonconvex, variable substitution and convex relaxation methods are needed to transform them into a form satisfying second-order cone programming. Then, the second-order convex programming (SCOP) method is used to solve the optimal operation problem. Therefore, we can let... , And by using the convex relaxation method, equations (13)-(15) and (21) can be linearized into equations (31)-(34).

[0041] (31) (32) (33) (34) Similarly, let , By combining the convex relaxation method, equations (17)-(19) and (22) are linearized into equations (35)-(39).

[0042] (35) (36) (37) (38) Therefore, the above lower-level operation model can be transformed into the following second-order cone lower-level programming model: (39) Substitute the optimal addressing scheme into the second-order cone lower-level planning model to solve the problem and output the optimal scheduling strategy. The upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system are iteratively optimized. The two layers interact through cost information until the iterative optimization termination condition is met, and the hierarchical planning scheme of the three-terminal low-voltage DC remote power supply system is output.

[0043] Simulation verification This application uses a 14-node low-voltage distribution network with a reference voltage of 0.38kV as the test system for simulation analysis. The network topology is as follows: Figure 2 As shown in the diagram. A schematic diagram of a three-terminal parallel low-voltage DC remote power supply system is shown below. Figure 3 As shown in Table 1.

[0044] Table 1 Example Parameters ; To verify the economy and effectiveness of the method proposed in this application, three schemes are set up for comparison in this embodiment. The schemes are as follows: Option 1: This is a blank control group. No parallel low-voltage DC remote power supply system is connected to the low-voltage AC power distribution test system, and no capacitors are installed. The optimized operating results are obtained using the Cplex solver.

[0045] Option 2: This is the traditional option, where the low-voltage AC power distribution test system only needs to install capacitor equipment at node 14 for operational optimization.

[0046] Option 3: This application proposes a method for planning a parallel low-voltage DC remote power supply system and using it to optimize the operation of the distribution network.

[0047] (1) Economic comparison of different planning schemes The above schemes were planned and solved, and the planning results, investment, maintenance and operating costs of each scheme are shown in Table 2.

[0048] Table 2. Comparison of Planning Results and Economic Efficiency of Different Schemes ; As shown in Table 2, although Scheme 2 (i.e., the traditional scheme) only incurs an investment cost of 28,100 yuan, its operating cost is reduced by only 6.58%, resulting in an economic benefit increase of 102,100 yuan / year. However, through the proposed hierarchical planning method, this application scheme installs a 100 kVA rectifier main unit and an 80 kVA inverter slave unit at nodes 2 and 13 respectively, with an intermediate DC line length of 840 meters. The investment cost is 34,400 yuan / year, which is a smaller increase in cost compared to Scheme 2, but the operating cost is reduced by 186,800 yuan / year (a decrease of 9.43%), resulting in an economic benefit of 152,400 yuan / year. Compared to the traditional scheme, this application scheme can effectively reduce investment and operating costs and achieve higher economic benefits.

[0049] The detailed operating costs of different schemes are shown in Table 3.

[0050] Table 3. Comparison of Operating Costs of Different Schemes ; As shown in Table 3, the electricity purchase costs of the three schemes are similar. Scheme 2 reduces the network loss cost by RMB 11,800 / year compared to Scheme 1, a decrease of 13.61%, while the network loss cost of the scheme in this application is reduced by RMB 40,500 / year, a decrease of as much as 46.71%. In addition, the scheme in this application has the lowest costs for electricity purchase, network loss, and voltage penalty, indicating that the parallel low-voltage DC remote power supply system is more economical than the traditional scheme for solving voltage over-limit problems.

[0051] (2) Validity analysis Table 4 Comparison of voltage quality indicators for different schemes ; Figure 4 The convergence curve of the genetic algorithm for hierarchical planning in this embodiment is given. As can be seen from the figure, the genetic algorithm and the hybrid algorithm of second-order cone programming in this application converge after the fourth iteration. Figure 5The voltage distribution diagrams for different schemes are presented. It can be seen that Scheme 1 exhibits voltage exceedance issues, while Scheme 2 effectively mitigates voltage exceedances through local reactive power compensation using capacitors, but suffers from significant overall voltage fluctuations. This implementation method, through optimization of the parallel low-voltage DC remote power supply system, completely resolves the voltage exceedance problem with minimal voltage fluctuations. Table 4 shows that without any voltage regulation measures, the average voltage qualification rate of the low-voltage distribution network is only 89.9%. After implementing voltage regulation measures in Scheme 2 and this application, the voltage qualification rate reaches 100%. However, in terms of voltage stability, the traditional scheme controls the voltage fluctuation standard deviation at 0.0236, only 16.61% higher than Scheme 1, while the proposed solution controls the voltage fluctuation standard deviation at 0.0167, 40.99% higher than Scheme 1, demonstrating a significant improvement in voltage stability.

[0052] In summary, the method proposed in this application presents an improved three-port low-voltage DC remote power supply system, expanding the voltage regulation range and enhancing voltage adjustment flexibility. Considering the power balance relationship between the rectifier, inverter, and DC line in the low-voltage DC remote power supply system, an investment and maintenance cost calculation model for the three-port low-voltage DC remote power supply system is established. The influence of the VSC control mode is taken into account, and its connection location, capacity, control mode, line length installation, and operating constraints are constructed to ensure the physical feasibility of the solution. A two-layer joint planning architecture is adopted, and the genetic algorithm is improved. The genetic algorithm and mathematical programming methods are combined to form a hybrid optimization algorithm, improving the ability to solve complex problems and enhancing computational convergence. Furthermore, by coordinating and optimizing the three-port low-voltage DC remote power supply system, the problem of user voltage exceeding limits is effectively solved, and investment and operating costs are reduced, improving economic efficiency.

[0053] In some embodiments, this application also provides a computer system including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0054] This application also provides a computer-readable storage medium for storing a computer program. This computer-readable storage medium can be applied to a computer device, and the computer program causes the computer device to execute the corresponding processes in the methods described above in the embodiments of this application; for brevity, further details are omitted here.

[0055] The above embodiments are preferred implementations of this application. In addition, this application can be implemented in other ways. Any obvious substitutions without departing from the concept of this technical solution are within the protection scope of this application.

[0056] To facilitate understanding by those skilled in the art of the improvements made by this application compared to the prior art, some of the accompanying drawings and descriptions have been simplified, and for clarity, some other elements have been omitted from this application. Those skilled in the art should realize that these omitted elements may also constitute the content of this application.

Claims

1. A hierarchical planning method for a three-terminal low-voltage DC remote power supply system for distribution network optimization, characterized in that, include: Collect basic data on low-voltage distribution network planning and operation; Based on the basic data of low-voltage distribution network planning and operation, an upper-level planning model for low-voltage DC remote power supply system is constructed with the objective function of minimizing the comprehensive investment, maintenance cost and lower-level feedback system operation cost of the low-voltage DC remote power supply system. Construct a lower-level optimization model for a low-voltage DC remote power supply system with the objective function of minimizing the power purchase cost, network loss cost, and voltage penalty cost of the distribution network; The upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system are solved iteratively to determine the hierarchical planning scheme of the three-terminal low-voltage DC remote power supply system.

2. The method according to claim 1, characterized in that, The basic data for low-voltage distribution network planning and operation includes the grid structure and load information of the low-voltage distribution network system. The grid structure of the low-voltage distribution network system includes the impedance information of each branch. The load information includes the active and reactive loads of each node in the low-voltage distribution network.

3. The method according to claim 2, characterized in that, The constraints of the upper-level planning model for the low-voltage DC remote power supply system include the following: VSC control mode constraints, access location and capacity constraints, and line length constraints.

4. The method according to claim 3, characterized in that, The constraints of the lower-level optimization model of the low-voltage DC remote power supply system include the following: Branch power flow constraints, node voltage constraints, and power constraints; wherein, the branch power flow constraints include AC power flow constraints and DC power flow constraints, and the power constraints include DC power balance constraints and AC power balance constraints.

5. The method according to claim 1, characterized in that, The process of iteratively solving the upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system to determine the hierarchical planning scheme of the three-terminal low-voltage DC remote power supply system includes: A hybrid optimization algorithm combining genetic algorithm and second-order cone programming is used to iteratively solve the upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system to determine the hierarchical planning scheme of the three-terminal low-voltage DC remote power supply system.

6. The method according to claim 5, characterized in that, The method employs a hybrid optimization algorithm combining genetic algorithm and second-order cone programming to iteratively solve the upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system, thereby determining the hierarchical planning scheme for the three-terminal low-voltage DC remote power supply system, including: A genetic algorithm is used to solve the upper-level planning model of the low-voltage DC remote power supply system and output the optimal addressing and capacity setting scheme. By employing variable substitution and convex relaxation, a second-order cone relaxation transformation is performed on the lower-level optimization model of the low-voltage DC remote power supply system to obtain a second-order cone lower-level programming model in convex optimization form. Substitute the optimal addressing and occupancy scheme into the second-order cone lower-level planning model for solution, and output the optimal scheduling strategy; The upper-level planning model and the lower-level optimization model of the low-voltage DC remote power supply system are iteratively optimized until the iterative optimization termination condition is met, and the hierarchical planning scheme of the three-terminal low-voltage DC remote power supply system is output.

7. The method according to claim 6, characterized in that, The iterative process of the genetic algorithm includes the design of decision variables and hybrid encoding, the construction and calculation of the fitness function, selection operation, crossover operation, and mutation operation; wherein, the selection operation adopts roulette wheel selection and elite retention selection; the crossover operation adopts categorized crossover; the binary part of the mutation operation adopts bit-flip mutation; the calculation result of the fitness function directly maps the optimization target achievement degree of the upper-level planning model of the low-voltage DC remote power supply system.

8. A computer system, characterized in that, include: Memory is used to store instructions that can be executed by the processor; A processor for executing the instructions to implement the method as described in any one of claims 1 to 7.

9. A computer-readable medium, characterized in that, The system contains computer program code that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images