Large power grid planning method based on adaptive improved genetic algorithm

Abstract

The invention relates to a large power grid planning method based on an adaptive improved genetic algorithm, and the method comprises the following steps: constructing a power grid topology model based on a power grid structure, and constructing a load flow calculation model and a load prediction model based on the power grid topology model; based on the load flow calculation model and the load prediction model, constructing a multi-objective optimization model by taking the cost of power grid planning, the reliability of a power grid system and the voltage stability as optimization objectives; setting a fitness function, and solving the multi-objective optimization model by adopting a self-adaptive improved genetic algorithm to obtain a power grid planning scheme; wherein the crossover rate and the mutation rate of the adaptive improved genetic algorithm are adaptively adjusted according to a fitness value, and dynamic scheduling optimization is performed on an optimization scheme generated in an iteration process by using a dynamic programming algorithm; according to the method, efficient planning can be realized in a complex power environment, the reliability and economical efficiency of power grid operation are remarkably improved, and the requirement of large-scale access of renewable energy sources in the future for power grid stability is met.

Description

Technical Field

[0001] The invention relates to a large power grid planning method based on an adaptive improved genetic algorithm, belonging to the technical field of power grid planning. Background Art

[0002] With the adjustment of the global energy structure and the large-scale application of renewable energy, grid planning and operation are facing unprecedented challenges. With the widespread access of distributed renewable energy such as wind power and photovoltaics to large power grids, the dynamics and complexity of the grid have increased significantly. Traditional grid planning technology is mainly aimed at optimizing centralized power generation and one-way power transmission, relying on static power flow calculation methods such as the Newton-Raphson method and linear programming. However, the access of distributed energy has changed this model. New energy generation has significant volatility and randomness, and its output power is affected by external factors such as weather conditions and day and night changes, resulting in increased uncertainty in the operation of the grid.

[0003] First, the intermittent nature of distributed renewable energy poses a severe challenge to the voltage and frequency stability of the power grid. When wind or photovoltaic power generation exceeds load demand, excess power will be transmitted back to the main grid, causing reverse power flow, which may cause voltage deviation, line overload and other problems. Traditional power flow calculation methods assume that the system is in a steady state and cannot reflect these dynamic changes in a timely manner, causing the calculation results to deviate from the actual situation.

[0004] Secondly, with the increase in the scale of new energy access, the structure and load distribution of the power grid have become more complex. The traditional static planning method shows low computational efficiency when dealing with large-scale complex power grids. Especially in the case of high penetration of new energy access, the computing time and resource requirements increase significantly, making it difficult to meet the needs of real-time planning and rapid response of the power grid.

[0005] In addition, although the existing dynamic planning has solved some timing problems to a certain extent, it still has problems with real-time performance and flexibility when facing the rapid fluctuations of new energy. Existing technologies also rely on a large amount of accurate input data, such as node loads, power generation equipment output, and line parameters. However, these data are often incomplete or inaccurate in practical applications, further limiting the effectiveness of planning methods. Summary of the invention

[0006] In order to solve the problems existing in the prior art, the present invention provides a large power grid planning method based on an adaptive improved genetic algorithm for improving dispatching efficiency.

[0007] In order to achieve the above object, the technical solution proposed by the present invention is: a large power grid planning method based on an adaptive improved genetic algorithm, characterized in that it includes the following steps:

[0008] Construct a power grid topology model based on the power grid structure, and construct a power flow calculation model and a load forecasting model based on the power grid topology model;

[0009] Based on the power flow calculation model and load forecasting model, a multi-objective optimization model is constructed with the cost of power grid planning, the reliability of the power grid system and the voltage stability as the optimization targets;

[0010] Set the fitness function and use the adaptive improved genetic algorithm to solve the multi-objective optimization model to obtain the power grid planning scheme;

[0011] The crossover rate and mutation rate of the adaptive improved genetic algorithm are adaptively adjusted according to the fitness value, and the optimization scheme generated in the iterative process is dynamically scheduled and optimized using a dynamic programming algorithm.

[0012] A further design of the above technical solution is as follows: the crossover rate adaptive adjustment formula is:

[0013]

[0014] Where: P cross (t) is the crossover rate of the tth generation; and are the minimum and maximum values ​​of the crossover rate respectively; f best is the highest fitness in the current population; f avg is the average fitness of the population; f worst is the minimum fitness;

[0015] The mutation rate adaptive adjustment formula is:

[0016]

[0017] in: and are the minimum and maximum values ​​of the mutation rate, respectively.

[0018] The formula for dynamic scheduling optimization is:

[0019]

[0020] Where: C gen (t) is the power generation cost at time t; C load (t) is the load adjustment cost at time t; T is the total number of scheduling time periods.

[0021] The power flow calculation model is:

[0022]

[0023] Among them, P i is the active power of node i, Vi and V j are the voltages at nodes i and j, G ij and B ij , is the admittance between nodes, θ i and θ j is the voltage angle.

[0024] The power flow calculation model is:

[0025] L(t)=L0+α·ΔT

[0026] Among them, L(t) represents the load at time t, L0 is the reference load, α is the load growth coefficient, and ΔT is the time increment.

[0027] The multi-objective optimization model is:

[0028] F(x)=w1×C 经济性 (x)+w2×C 可靠性 (x)+w3×C 电压稳定性 (x)

[0029]

[0030] Among them, C 经济性 (x) is the cost function of power grid planning, C 可靠性 (x) represents the reliability of the power grid system, C 电压稳定性 (x) represents voltage stability; w1, w2, w3 are weight coefficients of each target, L i is the predicted load demand from the load forecasting model, representing the power demand of node i; C i is the running cost of the node, P j is the power loss of line j from the power flow calculation model, C 损耗 is the cost per unit of power loss, R k is the reliability coefficient of node k, L 负荷 (t) is the predicted load demand from the load forecasting model, which indicates the load fluctuation at different time periods t. i is the voltage value of node i from the power flow calculation model, V 额定 is the rated voltage of the grid, which serves as the target value for stability.

[0031] The fitness function is:

[0032]

[0033] Among them, w1, w2, and w3 are the weights of each optimization objective.

[0034] The power grid topology model includes a node model, a line model, a power generation equipment model, a substation and transformer model, a load model, and a power flow direction and power flow model.

[0035] An electronic device comprises a processor, an input interface, an output interface and a memory, wherein the processor, the input interface, the output interface and the memory are interconnected, wherein the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the large power grid planning method based on the adaptive improved genetic algorithm as described above.

[0036] A computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the large power grid planning method based on the adaptive improved genetic algorithm as described above.

[0037] The beneficial effects of the present invention are:

[0038] (1) By introducing an adaptive genetic algorithm, the present invention can adaptively adjust the crossover rate and mutation rate, dynamically adapt to changes in the power grid operating environment, and optimize the allocation and scheduling of power grid resources. It can perform intelligent scheduling in a complex dynamic environment, optimize the operating status of the power grid, and improve the overall coordination and stability of the system. Compared with traditional methods, this technology can more flexibly deal with load fluctuations and the randomness of power generation, significantly improving the scheduling efficiency of power grid planning.

[0039] (2) The present invention uses a multi-objective optimization model to achieve a balance between economy, reliability and voltage stability, and constructs a highly adaptable and flexible power grid planning scheme to ensure the lowest power grid operation cost and stable operation. Traditional power grid planning methods usually focus on a single goal, while the method of the present invention can perform global optimization in a complex multi-objective environment, reducing the system operation cost while improving the safety and reliability of the power grid.

[0040] (3) In view of the uncertainty brought by the access of new energy, the present invention optimizes the power grid's ability to absorb renewable energy such as wind energy and photovoltaics through genetic algorithms. This technology can effectively reduce the phenomenon of wind and solar power abandonment, improve the utilization efficiency of new energy, promote the smooth operation of large-scale new energy access to the power grid, and help achieve the green transformation of the power grid.

[0041] (4) The present invention uses dynamic programming technology to make real-time adjustments according to the grid status in different time periods to ensure that the grid maintains stable operation under different load conditions. Compared with the existing static planning method, dynamic planning improves the adaptability of the grid to power generation and load changes and enhances the anti-interference ability of the system.

[0042] (5) The present invention greatly improves computing efficiency through distributed computing technology, especially in large-scale power grid environments, which can effectively reduce computing time and ensure the real-time response capability of power grid planning and scheduling. Compared with traditional methods, the present invention can more efficiently complete the planning and optimization of large-scale power grids. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A comparison diagram of actual power generation and predicted power generation in an example of the present invention;

[0044] Figure 2 This is a prediction error analysis diagram in an example of the present invention. DETAILED DESCRIPTION

[0045] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] Embodiment 1

[0047] The large power grid intelligent planning method of this embodiment includes the following steps:

[0048] Step 1: Build a grid topology model based on the grid structure.

[0049] The large power grid intelligent planning method of this embodiment builds a topological model of the power grid based on the basic structure of the power grid. The model includes nodes (representing power stations, substations or load centers) and lines connecting each node. The topological model of the power grid can describe the flow of electricity in the system:

[0050] 1) Node modeling

[0051] Each node in the power grid can represent a key equipment such as a power plant, substation, load center, etc. In power grid planning, the status of each node must be clear, including node voltage, node power and node type.

[0052] 2) Circuit modeling

[0053] The lines in the power grid are responsible for connecting different nodes. Electricity is transmitted from the generation node to the load node through the lines. When modeling each line, the line impedance, line capacity, line length and type need to be considered.

[0054] 3) Modeling of power generation equipment

[0055] Power generation equipment is the main source of electrical energy in the power grid. The modeling of power generation equipment mainly includes power generation type, rated power and power generation cost.

[0056] 4) Substation and transformer modeling

[0057] Substations are important hubs in power grids, responsible for changing voltage levels and dispatching power flows. The substation and its internal transformers need to specify transformer capacity, transformation ratio, and substation connection structure in the model.

[0058] 5) Load modeling

[0059] Load nodes represent end users or electrical equipment that consume electricity. The size and distribution of loads directly affect the operation of the power grid. Load modeling mainly considers load curves and load forecasts.

[0060] 6) Flow direction and power flow modeling

[0061] The core of the grid structure model lies in the flow and power flow direction of electricity. Under different load and power generation conditions, the flow direction of electricity and the amount of power transmission will change. The key to power flow modeling includes power flow direction and power balance.

[0062] Step 2: Construct a power flow calculation model and a load forecasting model based on the power grid topology model.

[0063] (1) Power flow calculation model;

[0064] The power flow calculation is an important part of this patent. It analyzes the operation status of the power grid by calculating the voltage of each node and the power flow of each line. The power flow calculation formula is as follows:

[0065]

[0066] Among them, P i is the active power of node i, V i and V j are the voltages at nodes i and j, G ij and B ij , is the admittance between nodes, θ i and θ j is the voltage angle. This power flow calculation model is used to evaluate the voltage level of each node in the power grid and the power transmission of the line.

[0067] (2) Load forecasting model;

[0068] In order to ensure accurate prediction of future loads in power grid planning, this patent uses historical data and power demand prediction models. By analyzing historical power consumption trends, combined with seasonal factors and load growth trends, the load demand of each node in the future is predicted. The load prediction formula is as follows:

[0069] L(t)=L0+α·ΔT (2)

[0070] Among them, L(t) represents the load at time t, L0 is the benchmark load, α is the load growth coefficient, and ΔT is the time increment. The model can plan the resource allocation of the power grid in advance and optimize the power supply based on the predicted load data.

[0071] Step 3: Based on the power flow calculation model and load forecasting model, a multi-objective optimization model is constructed with the cost of power grid planning, the reliability of the power grid system and voltage stability as optimization targets.

[0072] In order to achieve the economy, reliability and voltage stability of power grid planning, the present invention adopts a multi-objective optimization model, which comprehensively considers the following three optimization objectives:

[0073] F(x)=w1×C 经济性 (x)+w2×C 可靠性 (x)+w3×C 电压稳定性 (x)(3)

[0074]

[0075]

[0076] Among them, C 经济性 (x) is the cost function of power grid planning, C 可靠性 (x) represents the reliability of the power grid system, C 电压稳定性 (x) is used to quantify the voltage deviation of each node. w1, w2, w3 are the weight coefficients of each target, and their weight ratios can be adjusted according to actual needs. i is the predicted load demand from the load forecasting model, representing the power demand of node i. i is the operating cost of the node, which is determined by the node attributes in the grid structure model. j is the power loss of line j from the power flow calculation model, which is calculated based on the grid topology and power flow. 损耗 is the cost per unit of power loss, given by the grid structure and economic model. k is the reliability coefficient of node k, which is determined by the connection redundancy and backup power supply of the node in the power grid structure model. 负荷 (t) is the predicted load demand from the load forecasting model, which represents the load fluctuation at different time periods t and is used to evaluate the impact of load peak and valley fluctuations on grid reliability. i is the voltage value of node i from the power flow calculation model, which provides the real-time voltage status of each node. V 额定 is the rated voltage of the power grid, which is used as the target value for stability to ensure that the voltage deviation does not exceed the safety threshold. Through this model, the economy, stability and safety of power grid operation can be balanced at the same time.

[0077] Step 4: Set the fitness function and use the adaptive improved genetic algorithm to solve the multi-objective optimization model to obtain the power grid planning scheme; the crossover rate and mutation rate of the adaptive improved genetic algorithm are adaptively adjusted according to the fitness value, and the optimization scheme generated in the iterative process is dynamically dispatched and optimized using the dynamic programming algorithm; specifically:

[0078] In order to evaluate the pros and cons of each power grid planning scheme, this patent constructs a fitness function, which is used to comprehensively measure the realization of each optimization goal. The form of the fitness function is:

[0079]

[0080] Among them, w1, w2, and w3 are the weights of each optimization objective. Through the fitness function calculation, the pros and cons of each power grid planning scheme can be effectively evaluated to ensure that the optimal planning scheme is selected during the evolution process.

[0081] The genetic algorithm is used to solve the multi-objective optimization model, which includes the following steps:

[0082] Step 1: Population initialization;

[0083] In the population initialization stage, each individual x generated i is a candidate solution for the power grid planning scheme. Each individual contains the node and line attributes of the power grid, and its representation is as follows:

[0084] x i = {P i1 ,P i2 ,…,P in} (8)

[0085] Among them, P ij Represents individual x i Power allocation scheme on node j. Initial individual x i It is randomly generated, and the initial values ​​of node loads and line transmission capacity within the range can be determined based on the power grid structure model and load forecasting model.

[0086] Step 2: Fitness evaluation;

[0087] The fitness function F(x) of multi-objective optimization is used to measure the quality of each individual.

[0088] Step 3: Select individuals according to fitness values;

[0089] In the selection stage, the genetic algorithm selects individuals based on their fitness values. Individuals with higher fitness values ​​are more likely to be selected into the next generation. Common selection methods include roulette selection and tournament selection.

[0090] This example selects roulette wheel selection: In roulette wheel selection, individual x i The probability of being selected P(x i ) is determined by its fitness value, and the formula is:

[0091]

[0092] Among them, F(x i ) is the individual x i The fitness value is N, and N is the size of the population. Individuals with high fitness are given a higher selection rate, increasing their chances of being inherited in the next generation.

[0093] Elite retention strategy: In each generation of selection, the individuals with the highest fitness are directly retained to the next generation to ensure that excellent solutions will not be lost in the subsequent crossover and mutation processes.

[0094] Step 4: Adaptive crossover and mutation;

[0095] Traditional steps: During the crossover process, two parent individuals are randomly selected and two offspring individuals are generated through gene recombination. The crossover rate of the traditional crossover operation is fixed, which easily leads to a decrease in the diversity of the population; and the individual genes are randomly mutated to maintain the diversity of the population. The traditional mutation rate is also a fixed value.

[0096] In genetic algorithms, crossover rate and mutation rate are key parameters that affect population diversity and search efficiency. In order to avoid falling into a local optimal solution too early, the crossover rate and mutation rate in this embodiment will be adaptively adjusted according to the fitness value. The specific formula is as follows:

[0097] Adaptive crossover rate formula:

[0098]

[0099] Where: P cross (t) is the crossover rate of the tth generation; and are the minimum and maximum values ​​of the crossover rate respectively; f best is the highest fitness in the current population; f avg is the average fitness of the population; f worst is the minimum fitness.

[0100] Adaptive mutation rate formula:

[0101]

[0102] This formula is similar to the crossover rate, ensuring that the mutation rate is increased as the population converges, thereby increasing the diversity of the population.

[0103] Step 5: Dynamic scheduling optimization;

[0104] In order to ensure that the best individuals in each generation can be passed to the next generation, this embodiment adopts an elite retention strategy, that is, the individuals with the highest fitness in each generation are directly retained to ensure that the excellent solutions will not be lost during the evolution process, thereby accelerating the convergence of the algorithm and ensuring that the global optimal solution is finally found.

[0105] In the dynamic scheduling process, this patent combines the dynamic programming algorithm to adjust the power flow and load changes of the power grid in real time to ensure the stable operation of the power grid. The optimization formula for dynamic scheduling is:

[0106]

[0107] Where: C gen (t) is the power generation cost at time t; C load (t) is the load adjustment cost at time t; T is the total number of scheduling time periods.

[0108] The goal of this formula is to minimize the cost of power generation and load adjustment in each period to ensure the efficient use of new energy and the economy of the power grid.

[0109] Dynamic dispatch optimizes the operation of the power grid by predicting the load demand and renewable energy power generation in the next period and dynamically adjusting the power generation plan and load distribution.

[0110] Step 6: When the preset number of iterations is reached or the population fitness converges, the algorithm terminates and outputs the individual with the highest fitness as the final optimization solution.

[0111] The genetic algorithm of this embodiment ensures that the best individual at the time of termination has the highest fitness through the elite retention strategy.

[0112] In order to ensure that the best individuals in each generation can be passed to the next generation, the present invention adopts an elite retention strategy, that is, the individuals with the highest fitness in each generation are directly retained to ensure that the excellent solutions will not be lost during the evolution process, thereby accelerating the convergence of the algorithm and ensuring that the global optimal solution is finally found.

[0113] The elite retention strategy is a key step in the genetic algorithm to ensure that the best individuals in each generation will not be lost. The elite retention strategy is selected using the following formula:

[0114] E(t)=max(F(x1),F(x2),...,F(x n )) (11)

[0115] Where: E(t) is the elite individual of the tth generation; F(x i ) is the fitness of the ith individual.

[0116] The elite individuals in each generation are directly retained in the next generation population to ensure the transmission of excellent genes. Through this strategy, the loss of the optimal solution in the evolution process is avoided and the convergence speed is accelerated.

[0117] This embodiment ensures the diversity and optimization efficiency of the population during the evolution process by adaptively adjusting the crossover rate and mutation rate, thereby improving the global search capability of the genetic algorithm. At the same time, the crossover operation generates a new individual by exchanging some genes of two individuals; the mutation operation prevents the population from converging to a local optimal solution too early by randomly modifying some gene positions of the individual.

[0118] Embodiment 2

[0119] This embodiment provides an electronic device, including a processor, an input interface, an output interface and a memory, wherein the processor, the input interface, the output interface and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the large power grid planning method based on the adaptive improved genetic algorithm as in the above-mentioned embodiment 1.

[0120] This embodiment also provides a computer-readable storage medium, which stores a computer program. The computer program includes program instructions. When the program instructions are executed by a processor, the processor executes the large power grid planning method based on the adaptive improved genetic algorithm as described in the first embodiment above.

[0121] Application Examples

[0122] In order to verify the effectiveness of the large power grid intelligent planning model based on adaptive genetic algorithm proposed in this patent, this example verifies the accuracy and reliability of the model through simulation and data verification. The verification process includes building a test data set, performing simulation calculations, and performing comparative analysis through actual power grid operation data in a certain area in the Yangtze River Delta.

[0123] In order to simulate the actual operation of the power grid in the region, this example collects several sets of power grid operation data, including power generation capacity, load demand, and line transmission power. Assuming that the total power generation capacity of the power grid is 2000MW, the load demand fluctuates in different time periods, and the data is as follows:

[0124] Table 1 Data on power generation capacity, load demand and line transmission power of a certain local power grid during a certain period of time

[0125]

[0126] Using the adaptive genetic algorithm proposed in the above embodiment, we perform power grid flow calculation on the above data, and the calculation results are as follows:

[0127] Table 2: Actual power generation and predicted power generation and error data for a certain power grid period

[0128]

[0129] From the error analysis, it can be seen that the error between the power generation results predicted by the model and the actual power generation data is within 1.5%, indicating that the model has high accuracy.

[0130] To more clearly display the model verification results, this example draws two charts, showing the comparison between actual power generation and predicted power generation, as well as the prediction error.

[0131] from Figure 1 It can be seen that the change trend of the power generation predicted by the model is consistent with the actual power generation in each time period, indicating that the model can better predict the power generation of the power grid. Figure 2 The prediction errors are kept within a small range, which further verifies the reliability and stability of the model.

[0132] Through model verification, we conclude that the adaptive genetic algorithm proposed in this invention can effectively carry out intelligent planning of the power grid and predict the power grid operation status with high accuracy, which is suitable for practical application in the large power grid environment in the region.

[0133] The verification of the above simulation data shows that the method of the present invention is efficient and accurate, ensuring the balance between power generation and load scheduling, and can effectively improve the stability and anti-interference ability of the power grid when facing large-scale distributed energy access. By predicting and comparing data from multiple time periods, the results show that this method can effectively respond to power fluctuations under different load conditions, greatly reducing power generation prediction errors, and improving the economy and safety of the overall power grid; it can achieve efficient planning in a complex power environment, significantly improve the reliability and economy of power grid operation, and meet the future large-scale access of renewable energy to the grid stability requirements.

[0134] The technical solutions of the present invention are not limited to the above-mentioned embodiments, and any technical solutions obtained by equivalent replacement methods fall within the scope of protection required by the present invention.

Claims

1. A large power grid planning method based on an adaptive improved genetic algorithm, characterized in that: The steps include: Construct a power grid topology model based on the power grid structure, and construct a power flow calculation model and a load forecasting model based on the power grid topology model; Based on the power flow calculation model and load forecasting model, a multi-objective optimization model is constructed with the cost of power grid planning, the reliability of the power grid system and the voltage stability as the optimization targets; Set the fitness function and use the adaptive improved genetic algorithm to solve the multi-objective optimization model to obtain the power grid planning scheme; The crossover rate and mutation rate of the adaptive improved genetic algorithm are adaptively adjusted according to the fitness value, and the optimization scheme generated in the iterative process is dynamically scheduled and optimized using a dynamic programming algorithm.

2. The large power grid planning method based on the adaptive improved genetic algorithm according to claim 1 is characterized by: The crossover rate adaptive adjustment formula is: Where: P cross (t) is the crossover rate of the tth generation; and are the minimum and maximum values ​​of the crossover rate respectively; f best is the highest fitness in the current population; f avg is the average fitness of the population; f worst is the minimum fitness; The mutation rate adaptive adjustment formula is: in: and are the minimum and maximum values ​​of the mutation rate, respectively.

3. The large power grid planning method based on the adaptive improved genetic algorithm according to claim 2 is characterized by: The formula for dynamic scheduling optimization is: Where: C gen (t) is the power generation cost at time t; C load (t) is the load adjustment cost at time t; T is the total number of scheduling time periods.

4. The large power grid planning method based on the adaptive improved genetic algorithm according to claim 1 is characterized by: The power flow calculation model is: Among them, P i is the active power of node i, V i and V j are the voltages at nodes i and j, G ij and B ij , is the admittance between nodes, θ i and θ j is the voltage angle.

5. The large power grid planning method based on the adaptive improved genetic algorithm according to claim 4 is characterized by: The power flow calculation model is: L(t)=L0+α·ΔT Among them, L(t) represents the load at time t, L0 is the reference load, α is the load growth coefficient, and ΔT is the time increment.

6. The large power grid planning method based on the adaptive improved genetic algorithm according to claim 5 is characterized by: The multi-objective optimization model is: F(x)=w1×C 经济性 (x)+w2×C 可靠性 (x)+w3×C 电压稳定性 (x) Among them, C 经济性 (x) is the cost function of power grid planning, C 可靠性 (x) represents the reliability of the power grid system, C 电压稳定性 (x) represents voltage stability; w1, w2, w3 are weight coefficients of each target, L i is the predicted load demand from the load forecasting model, representing the power demand of node i; C i is the running cost of the node, P j is the power loss of line j from the power flow calculation model, C 损耗 is the cost per unit of power loss, R k is the reliability coefficient of node k, L 负荷 (t) is the predicted load demand from the load forecasting model, which indicates the load fluctuation at different time periods t. i is the voltage value of node i from the power flow calculation model, V 额定 is the rated voltage of the grid, which serves as the target value for stability.

7. The large power grid planning method based on the adaptive improved genetic algorithm according to claim 6 is characterized by: The fitness function is: Among them, w1, w2, and w3 are the weights of each optimization objective.

8. The large power grid planning method based on the adaptive improved genetic algorithm according to claim 7 is characterized by: The power grid topology model includes a node model, a line model, a power generation equipment model, a substation and transformer model, a load model, and a power flow direction and power flow model.

9. An electronic device, characterized in that: It includes a processor, an input interface, an output interface and a memory, which are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the large power grid planning method based on the adaptive improved genetic algorithm as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor, the processor executes the large power grid planning method based on the adaptive improved genetic algorithm as described in any one of claims 1 to 8.

Images