New energy project intelligent planning method and system based on large language model

Abstract

The invention discloses a new energy project intelligent planning method and system based on a large language model, and relates to the technical field of new energy planning, and the method comprises the steps: constructing a new energy project planning knowledge base; the new energy project planning knowledge base comprises basic data required by project planning and is used for providing data support for the optimization model; based on the training data in the new energy project planning knowledge base, performing field adaptability training on the large language model to obtain a special large language model in the new energy project planning field; and based on a project demand input by a user, analyzing project parameters through the special large language model, inputting an analysis result into the optimization model, and generating a planning scheme of a new energy project. According to the method, automatic conversion from a natural language demand to a professional planning scheme is realized, the accuracy of energy balance constraint is ensured through the topological structure diagram of the energy system, a flexible switching optimization target of economic benefits and environmental benefits is supported, and finally intelligent generation and evaluation of the planning scheme are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of new energy planning, and in particular to a method and system for intelligent planning of new energy projects based on a large language model. Background Art

[0002] Traditional new energy project planning mainly relies on expert experience and a single data analysis method, including technologies such as mixed integer linear programming (MILP) and energy system digital twins. Among them, the MILP method optimizes the energy system by building a mathematical model, but when dealing with complex multi-energy coupling systems, the model building process is cumbersome and difficult to flexibly adapt to the needs of different scenarios. Digital twin technology performs simulation analysis by establishing a digital mapping of the energy system, but it has limitations when dealing with unstructured project requirements and it is difficult to achieve automatic conversion from user natural language descriptions to professional parameters.

[0003] In recent years, large language models have made breakthrough progress in the field of natural language processing, providing a new technical path for new energy project planning. However, the existing large language models face two major challenges when applied in professional fields: first, the lack of in-depth understanding of energy system expertise makes it difficult to accurately analyze project requirements and give reliable planning suggestions; second, it is difficult to organically combine with traditional optimization models, and it is impossible to achieve end-to-end automation from demand understanding to solution optimization. In addition, in the planning of multi-energy coupling systems, existing technologies still have problems such as complex energy balance constraint construction and single optimization objectives, which limit the practicality and reliability of planning solutions. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the present invention provides a new energy project intelligent planning method and system based on a large language model, which can solve the problems mentioned in the background technology.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a method for intelligent planning of new energy projects based on a large language model, comprising: constructing a new energy project planning knowledge base; the new energy project planning knowledge base contains basic data required for project planning, which is used to provide data support for the optimization model; based on the training data in the new energy project planning knowledge base, the large language model is trained for domain adaptability to obtain a special large language model for the new energy project planning field; based on the project requirements input by the user, the project parameters are parsed by the special large language model, and the parsing results are input into the optimization model to generate a planning scheme for the new energy project.

[0007] As a preferred solution of the intelligent planning method for new energy projects based on a large language model described in the present invention, wherein: constructing a new energy project planning knowledge base includes the following steps: classifying project input data to establish a project geographic parameter library, an energy system parameter library, and a project economic parameter library; standardizing the data in the project geographic parameter library, the energy system parameter library, and the project economic parameter library to generate a standardized data set; extracting project planning digital mapping fields from the standardized data set to construct an energy project planning knowledge graph; and generating a parameter input library for the optimization model based on the energy project planning knowledge graph.

[0008] As a preferred solution of the intelligent planning method for new energy projects based on a large language model described in the present invention, wherein: extracting project planning digital mapping fields from the standardized data set to construct an energy project planning knowledge graph includes the following steps: establishing an entity library of energy input nodes, energy conversion equipment, and energy system loads, and generating an energy system topology relationship diagram; identifying energy flow relationships between entities from the standardized data set, and if there is a multi-energy coupling scenario, adding conversion relationships across energy categories; extracting attribute fields of each entity in the entity library, and if the entity belongs to the energy conversion equipment category, establishing an energy conversion efficiency matrix; constructing an energy project planning knowledge graph based on the energy system topology relationship diagram, energy category conversion relationships, and energy conversion efficiency matrix.

[0009] As a preferred solution of the method for intelligent planning of new energy projects based on a large language model described in the present invention, wherein: based on the training data in the new energy project planning knowledge base, the large language model is trained for domain adaptability to obtain a special large language model in the field of new energy project planning, including the following steps: based on the energy project planning knowledge graph and parameter input library, a new energy project planning dialogue corpus and a technical knowledge base are constructed; instruction templates and professional vocabulary mapping tables are extracted from the dialogue corpus and the technical knowledge base, and domain knowledge learning is performed on the open source large language model; an evaluation data set including the energy system topology relationship diagram and the energy conversion efficiency matrix is ​​constructed, and the large language model is optimized and trained to obtain a special large language model in the field of new energy project planning.

[0010] As a preferred solution of the intelligent planning method for new energy projects based on a large language model described in the present invention, wherein: an evaluation data set including the energy system topological relationship diagram and the energy conversion efficiency matrix is ​​constructed, and the large language model is optimized and trained to obtain a special large language model in the field of new energy project planning, including the following steps: constructing a multimodal evaluation data set including energy system topological relationships, equipment parameters and operation strategies; through an instruction supervised fine-tuning training method, if the input is a topological structure description, a standardized system mapping expression is output, and if the input is a parameter query, a normalized parameter matching result is output; a reward model is constructed based on the energy conversion efficiency matrix, and a penalty weight is imposed if the energy conversion path output by the model violates the laws of thermodynamics; a direct preference optimization method is used to finally train the model, and the evaluation result of the reward model is used as the optimization target to obtain a special large language model in the field of new energy project planning.

[0011] As a preferred solution of the intelligent planning method for new energy projects based on a large language model described in the present invention, generating a planning scheme for a new energy project includes the following steps: receiving a project requirement text input by a user, converting the natural language requirement into a structured parameter list through the dedicated large language model; inputting the structured parameter list into a mixed integer linear programming optimization model to generate a numerical solution including an energy equipment configuration scheme and an operation strategy; and generating a project planning text report based on the numerical solution, which includes an economic benefit analysis, an environmental benefit analysis, and a technical feasibility analysis.

[0012] As a preferred solution of the intelligent planning method for new energy projects based on a large language model described in the present invention, wherein: the structured parameter list is input into a mixed integer linear programming optimization model to generate a numerical solution including an energy equipment configuration plan and an operation strategy, including the following steps: the structured parameter list is classified and mapped according to energy input parameters, conversion equipment parameters, and load demand parameters; the objective function of the optimization model is established based on the energy conversion efficiency matrix, if the user selects the optimal economic benefit, the objective function of minimizing the total cost is set, if the user selects the optimal environmental benefit, the objective function of minimizing carbon emissions is set; an energy balance constraint equation group is constructed according to the energy system topology relationship diagram, and various equipment capacity constraints are set based on the parameter mapping results; a solver is called to calculate the optimization model, and the optimal numerical solution that meets the constraint conditions is output.

[0013] To further solve the above technical problems, the present invention provides the following technical solutions: A new energy project intelligent planning system based on a large language model, comprising: a construction module for constructing a new energy project planning knowledge base;

[0014] A training module, which performs domain adaptability training on the large language model based on the training data in the new energy project planning knowledge base to obtain a dedicated large language model in the new energy project planning field;

[0015] The planning module parses the project parameters through the dedicated large language model based on the project requirements input by the user, inputs the parsing results into the optimization model, and generates a planning scheme for the new energy project.

[0016] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the intelligent planning method for new energy projects based on a large language model as described above are implemented.

[0017] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the intelligent planning method for new energy projects based on a large language model as described above are implemented.

[0018] Beneficial effects of the present invention: The present invention combines a large language model with mixed integer linear programming technology to construct an intelligent new energy project planning system, realizing the automatic conversion from natural language requirements to professional planning schemes. The system uses a dedicated large language model to analyze project parameters, uses the energy conversion efficiency matrix to build a multi-energy coupling optimization model, ensures the accuracy of energy balance constraints through the energy system topology diagram, and supports the flexible switching optimization goals of economic benefits and environmental benefits, and finally realizes the intelligent generation and evaluation of planning schemes. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0020] Figure 1 This is a schematic diagram of the overall process of an intelligent planning method for new energy projects based on a large language model proposed by the present invention;

[0021] Figure 2 This is a computer equipment diagram in a new energy project intelligent planning method based on a large language model proposed by the present invention. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Example 1, reference Figure 1 , as an embodiment of the present invention, provides an intelligent planning method for new energy projects based on a large language model.

[0025] S1: Build a knowledge base for new energy project planning.

[0026] Specifically, the new energy project planning knowledge base contains the basic data required for project planning, which is used to provide data support for the optimization model.

[0027] S1.1: Classify the project input data to establish a project geographic parameter library, an energy system parameter library, and a project economic parameter library.

[0028] It should be noted that the project geographic parameter library contains data such as the topography and climate characteristics of the project site. These data are not only used for traditional feasibility analysis in the present invention, but more importantly, they provide geographic spatial semantic features for subsequent large language model training, so that the model can understand the impact of different geographical conditions on the energy system. The energy system parameter library stores the technical parameters of energy conversion equipment. In the present invention, a standardized processing method is innovatively adopted to unify the parameters of different energy categories into comparable dimensions. This processing method solves the problem that it is difficult to unify and compare the parameters of multiple energy categories in traditional energy planning.

[0029] S1.2: Standardize the data in the project geographic parameter database, energy system parameter database, and project economic parameter database to generate a standardized data set.

[0030] S1.3: Extract project planning digital mapping fields from standardized data sets and construct energy project planning knowledge graphs.

[0031] Specifically, establish a physical library of energy input nodes, energy conversion equipment, and energy system loads, and generate an energy system topology diagram;

[0032] Identify energy flow relationships between entities from standardized data sets, and add conversion relationships across energy categories if there are multi-energy coupling scenarios;

[0033] Extract the attribute fields of each entity in the entity library. If the entity belongs to the energy conversion equipment category, establish an energy conversion efficiency matrix;

[0034] Based on the energy system topology diagram, energy category conversion relationship and energy conversion efficiency matrix, a knowledge graph for energy project planning is constructed.

[0035] It should be noted that the energy system topology diagram uses the entity relationship mapping method to form a complete energy flow network by establishing the connection relationship between energy input nodes, conversion equipment and loads. Different from the traditional simple database storage, it constructs a knowledge network with semantic associations, which provides a basis for the large language model to understand the overall architecture of the energy system.

[0036] In the energy conversion efficiency matrix, rows: represent input energy types (such as electricity, natural gas, thermal energy, etc.); columns: represent output energy types (such as electricity, thermal energy, cold energy, etc.); matrix elements: corresponding energy conversion efficiency values. Efficiency data is obtained by extracting the rated efficiency of various energy conversion equipment from the equipment database, such as the COP value of heat pumps (electricity → thermal energy / cold energy conversion efficiency), the thermal efficiency of gas boilers (natural gas → thermal energy conversion efficiency), and the charging and discharging efficiency of energy storage devices (electricity → storage conversion efficiency of electric energy). In addition, the matrix filling rules include: 1. Direct conversion: the efficiency value of single-step energy conversion is directly filled in the corresponding position; 2. Non-convertible: for energy pairs without direct conversion channels, fill in 0 or null value; 3. Self-storage: the storage conversion efficiency of the same type of energy is filled in the diagonal. Here is an example of a simplified 3×3 matrix form:

[0037]

[0038] The elements in the matrix represent energy conversion efficiency, where: 0.95 represents electrical energy storage efficiency (taking losses into account); 3.5 represents heat pump heating COP; 3.0 represents heat pump cooling COP; 0.45 represents gas power generation efficiency; 0.85 represents gas boiler efficiency; 0.40 represents thermoelectric conversion efficiency; 0.90 represents thermal energy storage efficiency; and 0 represents no direct conversion channel.

[0039] S1.4: Generate parameter input library of optimization model based on energy project planning knowledge graph.

[0040] Preferably, step S1 solves the problem of knowledge representation and data integration in new energy project planning. Traditional new energy project planning mainly relies on expert experience and single data analysis, and it is difficult to effectively process and utilize a large amount of heterogeneous data. The present invention constructs a structured knowledge base system, and for the first time, uniformly standardizes and semantically associates geographic information, energy system parameters, and economic data. This processing method not only improves the efficiency of data utilization, but more importantly, provides a high-quality domain knowledge foundation for the training of large language models. In particular, in the knowledge graph construction link, by establishing an energy system topological relationship diagram and an energy conversion efficiency matrix, a semantic representation of multiple energy systems is achieved. This representation method enables the large language model to understand the internal relationships of complex energy systems, and then provide more accurate planning suggestions. Compared with the prior art, the knowledge base construction method of the present invention not only supports traditional data query and analysis, but also supports semantic understanding and reasoning. This ability shows significant advantages in processing users' natural language queries and multi-objective optimization problems.

[0041] S2: Based on the training data in the new energy project planning knowledge base, the large language model is trained for domain adaptability to obtain a dedicated large language model for the new energy project planning field.

[0042] S2.1: Based on the energy project planning knowledge graph and parameter input library, construct new energy project planning dialogue data and technical knowledge base.

[0043] S2.2: Extract instruction templates and professional vocabulary mapping tables from dialogue corpus and technical knowledge base, and perform domain knowledge learning on the open source large language model.

[0044] S2.3: Construct an evaluation data set including an energy system topology diagram and an energy conversion efficiency matrix, optimize and train the large language model, and obtain a dedicated large language model in the field of new energy project planning.

[0045] Specifically, a multimodal evaluation dataset including energy system topological relationships, equipment parameters, and operation strategies is constructed;

[0046] Through the instruction supervised fine-tuning training method, if the input is a topological structure description, the output is a standardized system mapping expression, and if the input is a parameter query, the output is a standardized parameter matching result;

[0047] A reward model is built based on the energy conversion efficiency matrix. If the energy conversion path output by the model violates the laws of thermodynamics, a penalty weight is imposed.

[0048] The direct preference optimization method is used to conduct the final training of the model, and the evaluation results of the reward model are used as the optimization target to obtain a dedicated large language model in the field of new energy project planning.

[0049] It should be noted that in S2, the instruction template refers to the conversion of the standard operating procedures in the new energy project planning process into structured prompt statements. This conversion enables the large language model to accurately understand the user's intention and provide a standardized response. The professional vocabulary mapping table establishes the correspondence between industry terms and model vocabulary. Especially when processing energy conversion equipment parameters, this mapping ensures that the model can accurately understand and use professional terms. In terms of model training, the present invention introduces a reward model based on the energy conversion efficiency matrix. This design is different from the traditional language model training method. It uses physical laws and engineering constraints as hard constraints for model training. By introducing the constraints of thermodynamic laws in the training process, it is ensured that the energy system solution output by the model always conforms to the laws of physics. This training method solves the problem that large language models are prone to hallucinations in professional fields, and significantly improves the reliability of the model in new energy project planning.

[0050] Step S2 solves the adaptability problem of large language models in the field of new energy project planning. Although traditional large language models have powerful natural language processing capabilities, they often have problems such as knowledge loss, concept confusion and violation of physical constraints in professional fields. The present invention achieves deep adaptation of the model in the field of new energy by constructing a special training corpus and an innovative training method. In particular, the energy conversion efficiency matrix is ​​introduced as an evaluation criterion during the model training process. This method not only ensures that the model output conforms to the laws of physics, but also evaluates the feasibility of the scheme based on actual engineering parameters. Compared with traditional methods, this training strategy enables the model to understand natural language descriptions and engineering parameters at the same time, while maintaining language comprehension capabilities, accurately grasping the professional knowledge of the energy system. Experiments show that the model trained in this way can not only understand the user's non-professional description when processing queries related to new energy project planning, but also provide solutions that strictly meet engineering standards.

[0051] S3: Based on the project requirements input by the user, the project parameters are parsed through a dedicated large language model, and the parsing results are input into the optimization model to generate a planning scheme for the new energy project.

[0052] S3.1: Receive project requirement text input by the user, and convert the natural language requirement into a structured parameter list through the dedicated large language model.

[0053] S3.2: Input the structured parameter list into a mixed integer linear programming optimization model to generate a numerical solution including energy equipment configuration scheme and operation strategy.

[0054] Specifically, the structured parameter list is classified and mapped according to energy input parameters, conversion equipment parameters, and load demand parameters;

[0055] The objective function of the optimization model is established based on the energy conversion efficiency matrix. If the user chooses the best economic benefit, the objective function of minimizing the total cost is set; if the user chooses the best environmental benefit, the objective function of minimizing carbon emissions is set;

[0056] Construct the energy balance constraint equations according to the energy system topology diagram, and set various equipment capacity constraints based on the parameter mapping results;

[0057] Call the solver to calculate the optimization model and output the optimal numerical solution that meets the constraints.

[0058] S3.3: Generate a project planning text report based on the numerical solution, which includes economic benefit analysis, environmental benefit analysis and technical feasibility analysis.

[0059] S3 organically combines the large language model with the mixed integer linear programming optimization model. In the parameter analysis stage, the dedicated large language model realizes the accurate conversion from unstructured natural language description to standardized parameter list, which solves the problem that users need to manually input a large number of professional parameters in traditional methods. In particular, in S3.2, the energy conversion efficiency matrix is ​​used to guide the construction of the optimization model, ensuring that the model can accurately reflect the coupling characteristics of the multi-energy system. Compared with the traditional single energy category optimization, this method can simultaneously consider the conversion relationship between multiple energy forms such as electricity, heat energy, and cold energy, which improves the practicality of the optimization results.

[0060] In the solution process, the physical feasibility of the optimization scheme is guaranteed by introducing energy balance constraints based on topological structures. This constraint construction method avoids the energy conservation violation problem commonly seen in traditional optimization models and improves the reliability of the planning scheme. In addition, the model supports flexible switching between economic and environmental benefits. By setting different objective functions, differentiated optimization schemes can be generated according to user needs. This multi-objective optimization capability is of great value in practical engineering applications.

[0061] Finally, in the planning solution generation phase, the system can automatically conduct multi-dimensional solution evaluation and generate professional planning reports. This automated solution evaluation and report generation mechanism has greatly improved the efficiency of new energy project planning and reduced the degree of dependence on professionals.

[0062] In summary, the present invention combines a large language model with mixed integer linear programming technology to build an intelligent new energy project planning system, realizing the automatic conversion from natural language requirements to professional planning schemes. The system uses a dedicated large language model to analyze project parameters, uses the energy conversion efficiency matrix to build a multi-energy coupling optimization model, ensures the accuracy of energy balance constraints through the energy system topology diagram, and supports the flexible switching optimization goals of economic benefits and environmental benefits, and finally realizes the intelligent generation and evaluation of planning schemes.

[0063] Embodiment 2 is an embodiment of the present invention, which provides a new energy project intelligent planning system based on a large language model, including:

[0064] Construction module, used to build a knowledge base for new energy project planning;

[0065] A training module, which performs domain-adaptive training on the large language model based on the training data in the new energy project planning knowledge base to obtain a dedicated large language model in the field of new energy project planning;

[0066] The planning module parses project parameters through a dedicated large language model based on the project requirements input by the user, inputs the parsing results into the optimization model, and generates a planning scheme for the new energy project.

[0067] Example 3, reference Figure 2 , is an embodiment of the present invention, which is different from the previous embodiment in that: if the function 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 this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0068] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0069] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0070] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0071] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A new energy project intelligent planning method based on a large language model, characterized in that: include: Build a knowledge base for new energy project planning; The new energy project planning knowledge base contains basic data required for project planning and is used to provide data support for the optimization model; Based on the training data in the new energy project planning knowledge base, the large language model is trained for domain adaptability to obtain a dedicated large language model for the new energy project planning field; Based on the project requirements input by the user, the project parameters are parsed by the dedicated large language model, and the parsing results are input into the optimization model to generate a planning scheme for the new energy project.

2. The method for intelligent planning of new energy projects based on a large language model as claimed in claim 1, characterized in that: Building a new energy project planning knowledge base includes the following steps: Classify the project input data to establish the project geographic parameter database, energy system parameter database, and project economic parameter database; Standardizing the data in the project geographic parameter library, energy system parameter library, and project economic parameter library to generate a standardized data set; Extracting project planning digital mapping fields from the standardized data set to construct an energy project planning knowledge graph; Based on the energy project planning knowledge graph, a parameter input library of the optimization model is generated.

3. The intelligent planning method for new energy projects based on a large language model as claimed in claim 2, characterized in that: Extracting the project planning digital mapping fields from the standardized data set and constructing the energy project planning knowledge graph includes the following steps: Establish a physical library of energy input nodes, energy conversion equipment, and energy system loads, and generate an energy system topology diagram; Identify energy flow relationships between entities from the standardized data set, and if there is a multi-energy coupling scenario, add conversion relationships across energy categories; Extracting the attribute fields of each entity in the entity library, and if the entity belongs to the energy conversion equipment category, establishing an energy conversion efficiency matrix; Based on the energy system topology diagram, energy category conversion relationship and energy conversion efficiency matrix, an energy project planning knowledge graph is constructed.

4. The intelligent planning method for new energy projects based on a large language model as claimed in claim 3, characterized in that: Based on the training data in the new energy project planning knowledge base, the large language model is trained for domain adaptability to obtain a dedicated large language model for the new energy project planning field, including the following steps: Based on the energy project planning knowledge graph and parameter input library, construct a new energy project planning dialogue data and technical knowledge base; Extracting instruction templates and professional vocabulary mapping tables from the dialogue corpus and technical knowledge base, and performing domain knowledge learning on the open source large language model; An evaluation data set including the energy system topology diagram and the energy conversion efficiency matrix is ​​constructed, and the large language model is optimized and trained to obtain a special large language model in the field of new energy project planning.

5. The intelligent planning method for new energy projects based on a large language model as claimed in claim 4, characterized in that: Constructing an evaluation data set including the energy system topology diagram and the energy conversion efficiency matrix, optimizing and training the large language model, and obtaining a dedicated large language model in the field of new energy project planning, including the following steps: Construct a multimodal evaluation dataset that includes energy system topology, equipment parameters, and operation strategies; Through the instruction supervised fine-tuning training method, if the input is a topological structure description, the output is a standardized system mapping expression, and if the input is a parameter query, the output is a standardized parameter matching result; A reward model is constructed based on the energy conversion efficiency matrix, and a penalty weight is imposed if the energy conversion path output by the model violates the laws of thermodynamics; The direct preference optimization method is used to finally train the model, and the evaluation results of the reward model are used as the optimization target to obtain a special large language model in the field of new energy project planning.

6. The method for intelligent planning of new energy projects based on a large language model as claimed in claim 5, characterized in that: Generating a planning scheme for a new energy project includes the following steps: Receiving a project requirement text input by a user, and converting the natural language requirement into a structured parameter list through the dedicated large language model; Inputting the structured parameter list into a mixed integer linear programming optimization model to generate a numerical solution including an energy equipment configuration scheme and an operation strategy; A project planning text report is generated based on the numerical solution, which includes economic benefit analysis, environmental benefit analysis and technical feasibility analysis.

7. The method for intelligent planning of new energy projects based on a large language model as claimed in claim 6, characterized in that: Inputting the structured parameter list into a mixed integer linear programming optimization model to generate a numerical solution including an energy equipment configuration scheme and an operation strategy comprises the following steps: Classify and map the structured parameter list according to energy input parameters, conversion equipment parameters, and load demand parameters; The objective function of the optimization model is established based on the energy conversion efficiency matrix. If the user chooses the best economic benefit, the objective function of minimizing the total cost is set; if the user chooses the best environmental benefit, the objective function of minimizing carbon emissions is set; Constructing an energy balance constraint equation group according to the energy system topology diagram, and setting various equipment capacity constraints based on the parameter mapping results; The solver is called to calculate the optimization model and output the optimal numerical solution that meets the constraint conditions.

8. A system using the large language model-based intelligent planning method for new energy projects as described in any one of claims 1 to 7, characterized in that: include: Construction module, used to build a knowledge base for new energy project planning; A training module, which performs domain adaptability training on the large language model based on the training data in the new energy project planning knowledge base to obtain a dedicated large language model in the new energy project planning field; The planning module parses the project parameters through the dedicated large language model based on the project requirements input by the user, inputs the parsing results into the optimization model, and generates a planning scheme for the new energy project.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the new energy project intelligent planning method based on a large language model described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent planning method for new energy projects based on a large language model described in any one of claims 1 to 7 are implemented.

Images