Power distribution network planning scheme generation method and device based on large model, terminal and medium

Abstract

The invention discloses a power distribution network planning scheme generation method and device based on a large model, a terminal and a medium, and relates to the technical field of power distribution networks. According to the scheme provided by the invention, through combination of constraint embedding cue words and a power distribution network planning scene, the constraint embedding cue words are constructed, natural languages and mathematical constraints are unified, and a power distribution network planning scheme is generated; therefore, cross-modal unified modeling of description capable of being understood by human and rules capable of being calculated by a machine is achieved, a prompt word set is embedded based on constructed constraints in an actual power distribution network planning scene, and corresponding mathematical constraints can be automatically matched and called to be used for optimization operation of a large model by inputting description of a natural language; therefore, a planning scheme meeting actual engineering requirements is generated, the tedious process that the planning requirements in a natural language form need to be manually converted into mathematical formulas is avoided, and the power distribution network planning work efficiency is improved.

Description

Technical Field

[0001] This application relates to the technical field of distribution networks, and particularly to a method, device, terminal and medium for generating a distribution network planning scheme based on a large model. Background Art

[0002] With the large-scale grid connection of renewable energy and the rapid development of diversified loads (such as electric vehicles and smart buildings), distribution network planning needs to cope with the challenges of high-dimensional uncertainty and multi-objective dynamic trade-off. Currently, the conventional planning methods mainly include traditional planning methods based on algorithms such as genetic algorithms and mixed integer linear programming, as well as AI optimization methods of artificial intelligence technologies (such as reinforcement learning and deep learning). However, these methods cannot directly write the planning requirements in the form of natural language into mathematical constraints through artificial means. This process is easily affected by subjective factors such as the technical level of the writing personnel, resulting in phenomena such as fuzzy constraint expressions and conflicting optimization goals, delaying the work progress, and further leading to the technical problem of low efficiency in the existing distribution network planning work. Summary of the Invention

[0003] This application provides a method, device, terminal and medium for generating a distribution network planning scheme based on a large model, which is used to solve the technical problem of low efficiency in the existing distribution network planning work.

[0004] To solve the above technical problem, the first aspect of this application provides a method for generating a distribution network planning scheme based on a large model, including:

[0005] Obtain distribution network planning sample data;

[0006] According to the distribution network planning sample data, combined with the natural language constraints and mathematical expression constraints associated with the distribution network planning constraints, construct a constraint embedding prompt word set;

[0007] Obtain distribution network planning requirement information, extract information features through a preset large model, and obtain the planning task target information corresponding to the distribution network planning requirement information;

[0008] According to the planning task target information, determine the corresponding target constraint embedding prompt word and the planning scheme optimization objective function, and generate a distribution network planning scheme through the inference operation of the large model according to the target constraint embedding prompt word and the planning scheme optimization objective function.

[0009] Preferably, the expression of the constraint embedding prompt word set is specifically:

[0010]

[0011] In the formula, is the identifier of the constraint embedding prompt word set, It is a sample of the distribution network planning scheme. It is a set of constraints. It is a constraint described in natural language. It is a constraint expressed by a mathematical formula. It is a weight coefficient.

[0012] Preferably, before generating the distribution network planning scheme according to the target constraint embedding prompt words and the planning scheme target information, in combination with the inference operation of the large model, it further includes:

[0013] According to the planning scheme target information, in combination with a preset loss function, the weight parameters of the large model are fine-tuned through the LoRA optimization method.

[0014] Preferably, the fine-tuning of the weight parameters of the large model specifically includes:

[0015] Decompose the weight parameter matrix of the large model into two low-rank matrices, where the first low-rank matrix is a matrix based on the Gaussian distribution, and the second low-rank matrix is a zero matrix;

[0016] Fix one of the low-rank matrices in turn, and perform iterative adjustment on the other low-rank matrix through the least squares solution method;

[0017] When the loss change reaches the preset convergence condition or the number of iterations reaches the preset maximum iteration number threshold, the updated weight parameter matrix is obtained according to the sum of the product of the two low-rank matrices and the weight parameter matrix.

[0018] Preferably, the loss function is:

[0019]

[0020]

[0021] In the formula, is the total loss coefficient, represents the task loss coefficient, is the sample of the distribution network planning scheme, is the predicted output value corresponding to the sample of the distribution network planning scheme, is the actual value corresponding to the sample of the distribution network planning scheme, and A and B are two low-rank matrices respectively.

[0022] Preferably, generating the distribution network planning scheme according to the target constraint embedding prompt words and the planning scheme target information through the inference operation of the large model includes:

[0023] Obtaining several solutions of the planning scheme according to the target constraint embedding prompt words and the planning scheme target information through the inference operation of the large model.

[0024] Through the logic of Pareto dominance relationship, optimize the solution of the planning scheme to obtain the Pareto optimal solution, and determine the distribution network planning scheme corresponding to the Pareto optimal solution.

[0025] Preferably, after obtaining the distribution network planning sample data, it further includes:

[0026] Preprocess the obtained distribution network planning sample data, where the preprocessing includes: data cleaning, normalization, and denoising.

[0027] The second aspect of this application provides a distribution network planning scheme generation device based on a large model, including:

[0028] A data acquisition unit for acquiring distribution network planning sample data;

[0029] A constraint prompt word set construction unit for constructing a constraint embedding prompt word set according to the distribution network planning sample data, in combination with the natural language constraints and mathematical expression constraints associated with the distribution network planning constraints;

[0030] A planning goal determination unit for obtaining distribution network planning requirement information, and extracting information features through a preset large model to obtain the planning task goal information corresponding to the distribution network planning requirement information;

[0031] A planning scheme generation unit for determining the corresponding target constraint embedding prompt words and a planning scheme optimization objective function according to the planning task goal information, and generating a distribution network planning scheme through the inference operation of the large model according to the target constraint embedding prompt words and the planning scheme optimization objective function.

[0032] The third aspect of this application provides a distribution network planning scheme generation terminal based on a large model, including: a memory and a processor;

[0033] The memory is used to store program code, and the program code is used to implement a distribution network planning scheme generation method provided in the first aspect of this application;

[0034] The processor is used to read and execute the program code.

[0035] The fourth aspect of this application provides a computer-readable storage medium, in which program code is stored, and the program code is used to be read and executed by a processor to implement a distribution network planning scheme generation method provided in the first aspect of this application.

[0036] From the above technical solutions, it can be seen that this application has the following advantages:

[0037] The solution provided by this application combines the constrained embedding prompt with the distribution network planning scenario. First, by constructing the constrained embedding prompt, natural language and mathematical constraints are unified, thereby achieving cross-modal unified modeling of human - understandable descriptions and machine - computable rules. In the actual distribution network planning scenario, based on the constructed set of constrained embedding prompts, by inputting a natural language description, the corresponding mathematical constraints can be automatically matched and called for the optimization operation of the large - model, so as to generate a planning solution that meets the actual engineering requirements, avoiding the cumbersome process of first manually converting the natural - language - form planning requirements into mathematical formulas, thus improving the efficiency of the distribution network planning work. Brief Description of the Drawings

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0039] Figure 1 It is a schematic flowchart of an embodiment of a method for generating a distribution network planning solution based on a large - model provided by this application.

[0040] Figure 2 It is an overall logical block diagram of an embodiment of a method for generating a distribution network planning solution based on a large - model provided by this application.

[0041] Figure 3 It is a schematic structural diagram of an embodiment of a device for generating a distribution network planning solution based on a large - model provided by this application.

[0042] Figure 4 It is a schematic structural diagram of an embodiment of a terminal for generating a distribution network planning solution based on a large - model provided by this application. Detailed Embodiments

[0043] The embodiments of this application provide a method, device, terminal, and medium for generating a distribution network planning solution based on a large - model, which are used to solve the technical problem of low efficiency in existing distribution network planning work.

[0044] To make the invention purpose, features, and advantages of this application more obvious and understandable, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the drawings in the embodiments of this application. Obviously, the embodiments described below are only some embodiments of this application, rather than all embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application.

[0045] First, a detailed description of an embodiment of a method for generating a distribution network planning scheme based on a large model provided by this application is as follows:

[0046] Please refer to Figure 1 and Figure 2 , an embodiment of a method for generating a distribution network planning scheme based on a large model provided by this application includes:

[0047] Step 101, obtain distribution network planning sample data;

[0048] It should be noted that first, collect distribution network-related data from different fields, mainly including: power system operation data, historical planning data, geospatial data (GIS), environmental data, etc. The data sources can include power companies, sensors, satellite remote sensing, weather stations, etc.

[0049] Furthermore, after obtaining the data in step 101 and before executing step 102, the following steps can be further appended:

[0050] 1011. Preprocess the obtained distribution network planning sample data;

[0051] It should be noted that preprocess the collected original distribution network planning sample data , including operations such as data cleaning, normalization, and denoising. Among them, the normalization process standardizes each data using the Z-Score method to ensure the consistency and comparability of the data.

[0052]

[0053] In the formula, represents the th type of variable data from the th field, represents the average value of the th type of variable data from the th field, represents the variance of the th type of variable data from the th field, represents the variable data after standardization.

[0054] Step 102, according to the distribution network planning sample data, combine the natural language constraints and mathematical expression constraints associated with the distribution network planning constraints to construct a constraint embedding prompt word set;

[0055] It should be noted that based on the collected data, a constraint embedding prompt set engineering is constructed. Among them, the constraint embedding prompt mentioned in this embodiment is a technical tool that structurally encodes multi-dimensional constraints such as physics, economy, and policy in the distribution network planning into a combination of natural language and mathematical formulas, and uniformly encodes the multi-dimensional constraints such as physics, economy, and policy in the distribution network planning into a structured input, including a natural language description that can be understood by humans and a mathematical formula that can be calculated by machines. Its core purpose is to achieve low data dependence through descriptions that can be understood by humans and rules that can be calculated by machines. The expression of the constraint embedding prompt is as follows:

[0056]

[0057] In the formula, is the identifier of the constraint embedding prompt set, is a sample of the distribution network planning scheme, is a set of constraints, is the natural language description constraint, is the mathematical formula expression constraint, is the weight coefficient, and the weight ratio of the formula can be dynamically adjusted as needed.

[0058] Step 103, obtain the distribution network planning demand information, and extract the information features through a preset large model to obtain the planning task target information corresponding to the distribution network planning demand information;

[0059] It should be noted that by obtaining the actual distribution network planning demand information and extracting the information features through a preset large model, the key planning information, that is, the planning task target information, is extracted.

[0060] Among them, when using LoRA (Low-Rank Adaptation) for large model fine-tuning, the loss function can be set as:

[0061]

[0062]

[0063] Among them, represents the Frobenius norm of the matrix. The first term represents the task loss, is the task input to the large model, is the predicted value output by the large model for task X, is the actual value of task X. The second term represents the regularization term to prevent overfitting. By minimizing this loss function , the optimal fine-tuning strategy can be obtained.

[0064] Before executing step 104, it also includes:

[0065] Step 1031: According to the target information of the planning scheme, combined with a preset loss function, fine-tune the weight parameters of the large model through the LoRA optimization method.

[0066] More specifically, the fine-tuning of the weight parameters of the large model specifically includes:

[0067] Decompose the weight parameter matrix of the large model into two low-rank matrices, where the first low-rank matrix is a matrix based on the Gaussian distribution, and the second low-rank matrix is a zero matrix;

[0068] Fix one of the low-rank matrices in turn, and iteratively adjust the other low-rank matrix through the least squares solution method;

[0069] When the loss change reaches the preset convergence condition or the number of iterations reaches the preset maximum iteration threshold, obtain the updated weight parameter matrix according to the sum of the product of the two low-rank matrices and the weight parameter matrix.

[0070] The specific step process is as follows:

[0071] a. Parameter initialization: Assume that the weight matrix of the pre-trained model is , where is the input dimension, is the output dimension. Define two low-rank matrices , where is the low-rank dimension.

[0072] Random initialization: Matrix is randomly initialized using the Gaussian distribution (mean is 0, standard deviation is , and matrix is initialized as a zero matrix.

[0073]

[0074] Domain knowledge initialization: If the target task is related to the pre-training task, and can be initialized through partial information of the pre-training weights.

[0075] b. Iterative optimization: Repeat the following steps until convergence: Fix , update : According to the current estimate, update the factor matrix F by solving the least squares problem to minimize the objective function.

[0076]

[0077] Fix , update :

[0078]

[0079] where is the learning rate, generally set to ~ , , is the and gradient of

[0080] c. Convergence judgment: Judge whether the objective function reaches the convergence condition: the continuous loss change or the relative loss change rate or reaches the maximum number of iterations.

[0081] d. Output result: Finally, obtain the updated weight matrix :

[0082]

[0083] Retain most of the parameters of the pre-trained model, only replace or add and for target task inference. Support dynamic switching of and for different tasks, realize multi-task adaptation, so it is also possible to determine the corresponding and for different task types based on different planning task samples in advance, so as to be directly reused when actually facing the same distribution network planning task, and further improve the efficiency of the planning work.

[0084] LoRA fine-tuning can significantly reduce the computational resource consumption in power grid planning. Only by adjusting key parameters can it adapt to different scenarios, ensure that the planning scheme is both flexible and innovative and meets the actual engineering constraints, and avoid decision-making biases caused by over-reliance on data or experience.

[0085] Step 104: According to the planning task objective information, determine the corresponding target constraint embedding prompt words and the planning scheme optimization objective function, so as to generate a distribution network planning scheme through the inference operation of the large model based on the target constraint embedding prompt words and the planning scheme optimization objective function.

[0086] It should be noted that according to the planning task objective information obtained in the previous steps, the corresponding objective function for optimizing the planning scheme is determined, and combined with the constraint embedding prompt set project constructed in the previous steps, the objective constraint embedding prompts associated with the planning task objective information are matched. For example, the planning requirements proposed in the distribution network planning document obtained include three optimization objectives: total investment cost, average power supply availability rate, and carbon emissions. Based on these three optimization objectives, the overall optimization objective function can be initially determined. An example of the optimization objective function is as follows:

[0087]

[0088] In the formula, represents the planning scheme, represents different sub-objective functions. is the total investment cost, is the purchase cost per unit capacity of the equipment, is the annual operation and maintenance cost per unit of the equipment; is the operating time of the equipment, is the average power supply availability rate, represents the carbon emissions, is the carbon emission intensity per unit of power generation, is the carbon emission per unit capacity for construction, represents the power generation, represents the construction capacity. represents the weight of the sub-objective.

[0089] Then, combined with the pre-constructed constraint embedding prompt set project, the constraint embedding prompts corresponding to these three optimization objectives are matched, so as to use the mathematical constraints in the constraint embedding prompts as the constraint conditions associated with the objective function, thereby forming a complete optimization model equation including the objective function and the constraint conditions. Based on this optimization model equation, using the operation mechanism of the large model, a distribution network planning scheme is generated.

[0090] Furthermore, when the objective function constructed in step 104 is a multi-objective optimization function as provided in the above example, multi-objective optimization can be performed through the Pareto dominance relationship. Among them, the Pareto dominance relationship is the core concept in multi-objective optimization and is used to compare the advantages and disadvantages of two solutions. Its definition is: solution dominates solution (denoted as ), if and only if is not inferior to in all objectives (i.e., ), and is strictly better than in at least one objective (i.e., ), , ). This relationship ensures the overall superiority of the solution through dual conditions and is the basis for screening the Pareto optimal solution. By introducing the Pareto dominance relationship, the planning model can screen out the non-dominated solution set, namely the Pareto frontier, thereby providing decision makers with a series of optimal trade-off solutions, which helps to improve the execution efficiency of distribution network planning.

[0091] Furthermore, after generating the distribution network planning scheme, the generated planning scheme can also be evaluated by flow calculation simulation to determine whether it meets the actual engineering feasibility. For the schemes that do not meet the conditions, they are returned to the large model for re-optimization, so that the entire process forms a generation-feedback closed-loop optimization.

[0092] The embodiment of the present application proposes a mathematical-linguistic hybrid expression method based on constraint embedding, realizes seamless compatibility between physical rules and large language models (LLMs) through constraint embedding prompt word engineering, realizes cross-modal unified modeling of natural language descriptions understandable to humans and mathematical rules computable by machines, and based on the constructed constraint embedding prompt word set, can guide the pre-trained large model through natural language prompts to generate planning solutions that meet actual engineering needs through natural language constraints, avoiding the tedious process of manually converting natural language planning requirements into mathematical formulas, and ensuring the rationality and accuracy of the model generation results under the constraints of physical rules. At the same time, the low-rank adaptive multi-objective optimization strategy is adopted, combined with the LoRA fine-tuning technology and the Pareto ranking method, which significantly reduces the computational complexity, while taking into account the efficiency and accuracy of multi-objective optimization, and improving the overall performance of the model.

[0093] The above is a detailed description of an embodiment of a distribution network planning scheme generating method based on a large model provided in the present application. The following is a detailed description of an embodiment of a distribution network planning scheme generating device based on a large model provided in the present application.

[0094] See also Figure 3 , the present application provides an embodiment of a distribution network planning scheme generation device based on a large model, comprising:

[0095] The data acquisition unit 201 is used to acquire distribution network planning sample data;

[0096] The constraint prompt word set construction unit 202 is used to construct a constraint embedding prompt word set according to the distribution network planning sample data and in combination with the natural language constraints and mathematical expression constraints associated with the distribution network planning constraints;

[0097] The planning target determination unit 203 is used to obtain the distribution network planning demand information, extract information features through a preset large model, and obtain the planning task target information corresponding to the distribution network planning demand information;

[0098] The planning scheme generation unit 204 is configured to determine the corresponding target constraint embedding prompt and the planning scheme optimization objective function according to the planning task target information, so as to generate a distribution network planning scheme through the inference operation of the large model based on the target constraint embedding prompt and the planning scheme optimization objective function.

[0099] As Figure 4 shown, an embodiment of a distribution network planning scheme generation terminal based on a large model provided by the present application. The implementation types of the terminal include, but are not limited to: personal computers, industrial computers, servers, and embedded intelligent devices. The main components of the terminal include: a memory 33 and a processor 31. Among them, the memory 33 and the processor 31 can be connected through a communication bus 34;

[0100] The memory 33 is used to store program codes, and the program codes are used to implement a method for generating a distribution network planning scheme based on a large model as provided in the above embodiment;

[0101] The processor 31 is used to read and execute the program codes.

[0102] An embodiment of a computer-readable storage medium provided by the present application. The computer-readable storage medium stores program codes, and the program codes are used to be read and executed by a processor to implement a method for generating a distribution network planning scheme based on a large model as provided in the above embodiment.

[0103] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described terminal, device, and unit can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0104] In several embodiments provided by the present application, it should be understood that the disclosed terminal, device, and method can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, and the indirect coupling or communication connection of the device or unit may be in an electrical, mechanical, or other form.

[0105] In the description of this application and the above-mentioned accompanying drawings, terms such as "first", "second", "third", "fourth", etc. (if any) are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of this application described herein, for example, can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that comprises a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0106] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects and indicates that three relationships can exist. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist simultaneously. Here, A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one (one)" or a similar expression thereof refers to any combination of these items, including any combination of single items (ones) or plural items (ones). For example, at least one (one) of a, b, or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0107] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0108] In addition, in each embodiment of the present invention, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0109] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0110] As described above, the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of various embodiments of the present application.

Claims

1. A method for generating a distribution network planning scheme based on a large model, characterized in that Including: Obtain the sample data of the distribution network planning; According to the sample data of the distribution network planning, combined with the natural language constraints and mathematical expression constraints associated with the distribution network planning constraints, construct a constraint embedding prompt set; Obtain the distribution network planning requirement information, extract the information features through a preset large model, and obtain the planning task target information corresponding to the distribution network planning requirement information; According to the planning task target information, determine the corresponding target constraint embedding prompt and the planning scheme optimization objective function, so as to generate a distribution network planning scheme through the inference operation of the large model according to the target constraint embedding prompt and the planning scheme optimization objective function.

2. The method for generating a distribution network planning scheme based on a large model according to claim 1, wherein The expression of the constraint embedding prompt set is specifically: ； In the formula, is the identifier of the constraint embedding prompt set, is a sample of the distribution network planning scheme, is a set of constraints, is a natural language description constraint, is a constraint expressed by a mathematical formula, is the weight coefficient.

3. The method for generating a distribution network planning scheme based on a large model according to claim 1, wherein Before generating a distribution network planning scheme according to the target constraint embedding prompt and the planning scheme target information, combined with the inference operation of the large model, it further includes: According to the planning scheme target information, combined with a preset loss function, fine-tune the weight parameters of the large model through the LoRA optimization method.

4. The method for generating a distribution network planning scheme based on a large model according to claim 3, wherein The specific process of fine-tuning the weight parameters of the large model includes: Decompose the weight parameter matrix of the large model into two low-rank matrices, where the first low-rank matrix is a matrix based on the Gaussian distribution, and the second low-rank matrix is a zero matrix; Fix one of the low-rank matrices in turn, and iteratively adjust the other low-rank matrix through the least squares solution method; When the loss change reaches the preset convergence condition or the number of iterations reaches the preset maximum iteration threshold, obtain the updated weight parameter matrix according to the sum of the product of the two low-rank matrices and the weight parameter matrix.

5. The method for generating a distribution network planning scheme based on a large model according to claim 3, wherein, The loss function is: ； ； In the formula, is the total loss coefficient, represents the task loss coefficient, is the sample of the distribution network planning scheme, is the predicted output value corresponding to the sample of the distribution network planning scheme, is the actual value corresponding to the sample of the distribution network planning scheme, and A and B are two low-rank matrices respectively.

6. A method for generating a distribution network planning scheme based on a large model according to claim 1, characterized in that The process of generating a distribution network planning scheme according to the target constraint embedding prompt and the planning scheme target information through the inference operation of the large model includes: According to the target constraint embedding prompt and the planning scheme target information, obtain a number of planning scheme solutions through the inference operation of the large model; Through the logic of the Pareto dominance relationship, optimize the planning scheme solutions to obtain the Pareto optimal solution, and determine the distribution network planning scheme corresponding to the Pareto optimal solution.

7. A method for generating a distribution network planning scheme based on a large model according to claim 1, characterized in that, After obtaining the sample data of the distribution network planning, it further includes: Preprocess the obtained sample data of the distribution network planning, where the preprocessing includes: data cleaning, normalization, and denoising.

8. A device for generating a distribution network planning scheme based on a large model, characterized in that, Including: A data acquisition unit for obtaining the sample data of the distribution network planning; A constraint prompt set construction unit for constructing a constraint embedding prompt set according to the sample data of the distribution network planning, combined with the natural language constraints and mathematical expression constraints associated with the distribution network planning constraints; A planning target determination unit for obtaining the distribution network planning requirement information, extracting the information features through a preset large model, and obtaining the planning task target information corresponding to the distribution network planning requirement information; A planning scheme generation unit, configured to determine corresponding target constraint embedding prompt words and a planning scheme optimization objective function according to the planning task target information, so as to generate a distribution network planning scheme through the inference operation of the large model according to the target constraint embedding prompt words and the planning scheme optimization objective function.

9. A terminal for generating a distribution network planning scheme based on a large model, characterized in that, It includes: A memory and a processor; The memory is used to store program codes, and the program codes are used to implement a method for generating a distribution network planning scheme based on a large model according to any one of claims 1 to 7; The processor is used to read and execute the program codes.

10. A computer-readable storage medium, characterized in that, Program codes are stored in the computer-readable storage medium and are used to be read and executed by a processor to implement a method for generating a distribution network planning scheme based on a large model according to any one of claims 1 to 7.

Images