A feasibility study report automatic generation system and method based on large model

Abstract

The present invention discloses a system and method for automatically generating a feasibility study report based on a large model, and belongs to the field of artificial intelligence. The present invention adds a cache check block and sets cache check blocks of different version levels, and realizes fast search, update, deletion and addition operations according to the cache check block of the corresponding version level. By setting the cache check block version downgrade mode, the operation complexity of the cache data is reduced, and the operation speed of the system is greatly improved. A pre-cache area is added to the project data buffer, and the system performance is greatly improved by combining content perception, reducing response delays and improving cache hit rates. By preloading more data into the cache, the frequency of directly reading data from the database is reduced, the burden on the database is reduced, the life of the database is extended, and the overall maintenance cost of the system is reduced. It not only improves the user experience, but also reduces the network load and bandwidth costs, and brings operating space for the utilization of network resources.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a system and method for automatically generating a feasibility study report based on a large model. Background Art

[0002] The feasibility study report provides a data basis for investing, launching or abandoning a project by conducting a comprehensive analysis of the project's market demand, technical feasibility, financial feasibility, legal and environmental aspects, thereby reducing the blindness and arbitrariness of decision-making. Through scientific data analysis and data evaluation, it effectively identifies and predicts potential risks, provides guidance for the specific implementation of the project, and ensures that the project plan is carried out normally to achieve the expected goals. It can also demonstrate the potential benefits and feasibility of the project, attract the attention of investors and financing institutions, and enhance the confidence of investors and financing institutions by displaying the highlights, competitive advantages, and market prospects of the project through data. The feasibility study report can not only increase the probability of project success, but also effectively reduce project risks.

[0003] The feasibility study report covers all aspects of the project, including market analysis, technical analysis, organization and management, resource allocation, implementation plan, cost estimation and budget, investment benefit analysis, risk analysis and management, legal and policy analysis, environmental impact assessment, etc. The data information is huge. When the project data information is added, modified or deleted, resulting in abnormal data information, the generated feasibility study report may contain erroneous information or omit important data, thus affecting its accuracy and reliability. Abnormal data update requires additional time and resources to repair and regenerate the report, and the original data cannot be restored in time, thus affecting the project cycle. Summary of the invention

[0004] The purpose of the present invention is to provide a system and method for automatically generating a feasibility study report based on a large model, so as to solve the problems raised in the above-mentioned background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solution: a method for automatically generating a feasibility study report based on a large model.

[0006] The method for automatically generating a feasibility study report includes the following steps:

[0007] Step S1, establish a project data buffer, obtain data information corresponding to the project, store it in the project data buffer after data cleaning and de-redundancy processing, set a cache check block, the cache check block includes a check block header and a check block footer, add a check block header to the table header according to the data in the project data buffer, add a check block footer to the table footer, and the cache check block follows the non-nested marking rule;

[0008] Step S1-1, establish a project data buffer, and initialize the size and storage structure of the buffer according to the project data;

[0009] Step S1-2, obtaining data information corresponding to the project, and storing it in the project data buffer according to the project data classification after data cleaning and deduplication processing;

[0010] Step S1-3, setting a cache check block, the cache check block includes a check block header for identifying a data table header and a check block tail for identifying a data table tail, and the cache check block follows a non-nested marking rule;

[0011] Step S1-4: Use the cache check block to identify the data in the project data buffer.

[0012] When reading in project data for the first time, a project data buffer is first established, and the project data is read in after initialization. After data cleaning and redundancy removal, it is stored in the corresponding tables in the database. When accessing the data again, the data from the corresponding table in the database is stored in the project data buffer, and the data in the project data buffer is synchronized to the feasibility study report, and the data in the project data buffer is synchronized to the corresponding table in the database through asynchronous processing to complete the data access interaction.

[0013] Step S2, set cache verification monitoring to monitor and obtain the changed data in the project data buffer in real time. When the data in the project data buffer is changed, the cache verification monitoring will upgrade the corresponding cache verification block to mark the changed data, and mark the unchanged data with the original version cache verification block;

[0014] Step S2-1, set cache verification monitoring, and monitor and obtain the changed data in the project data buffer in real time;

[0015] Step S2-2, extract the changed data, and upgrade the corresponding cache check block through cache check monitoring, and the upgraded cache check block is used to identify the changed data;

[0016] Step S2-3: The unchanged data is marked with the original version cache check block, following the non-nested marking rule;

[0017] Step S2-4: extract the cache check block data of the version change, update the data in the project data buffer to the feasibility study report, and then update it to the database.

[0018] When the data changes, the corresponding cache check block version in the project data buffer is obtained, and the data in the feasibility study report is found according to the version. After the data in the project data buffer is updated to the feasibility study report, the data is asynchronously updated to the database to complete the data operation.

[0019] Step S3, setting a version downgrade block and a downgrade threshold, when the number of data changes in the project data buffer exceeds the downgrade threshold, the version downgrade block downgrades the versions of all cache check blocks;

[0020] Step S4: Analyze and extract the project type and the covered industry information from the project data buffer, and dynamically generate an adaptive personalized feasibility study report template;

[0021] Step S4-1, extract the project type and the covered industry information through the project data buffer analysis, extract the involved industry information from the project data and objectives, determine the industry field covered by the project, set the industry classification as the first-level classification, and set the project data as the second-level classification;

[0022] Industry classification includes: manufacturing, agriculture, forestry, fishery and animal husbandry, mining, construction, energy and utilities, information technology and communications, finance and insurance, real estate, transportation and logistics, etc.

[0023] Project feasibility data include: market data, technical data, financial data, economic data, environmental data, risk data, laws and policies, resource data and implementation plan;

[0024] Step S4-2: Match the adaptation template in the template library according to the project type and industry information of the first-level classification, and adjust the template according to the second-level classification information to generate the final feasibility study report template.

[0025] Step S5: According to the data in the project data buffer and the personalized feasibility study report template, the project data is read out from the project data buffer area, and the project data is matched with the feasibility study report template after data mapping processing to generate a feasibility study report with chapters and contents;

[0026] Step S5-1, loading the personalized feasibility study report template generated by step S4, and reading the project data in the project data buffer area;

[0027] Step S5-2: Render the data to the corresponding position in the feasibility study report template through data matching and data mapping, and generate the corresponding chapters and contents of the feasibility study report.

[0028] Step S6, add a pre-cache area to the project data buffer, obtain the changed project data, predict the project data that needs to be accessed through content perception, and add it to the pre-cache area in advance. When the project data buffer is updated as a whole, repeat the adding method of step S1 to add a cache check block. When the acquired project data information changes, the data in the project data buffer is changed synchronously, and the data in the feasibility study report is updated synchronously through the cache check block version level;

[0029] Furthermore, set the version downgrade block and the downgrade threshold, and set the downgrade threshold to δ max , when the number of data changes in the project data buffer exceeds the downgrade threshold δ max When the version downgrade block is used, all cache check blocks are downgraded. The level of each version downgrade is 1. The initial level of the original version cache check block is 0. When the version level of the cache check block is 0, it will no longer be downgraded. Two adjacent cache check blocks of the same version in a database table are merged by default.

[0030] Further, a pre-cache area is added to the project data buffer, the changed project data is obtained, and the project data to be accessed is predicted by content perception, wherein the content perception prediction calculation includes cosine similarity calculation and Euclidean distance calculation;

[0031] The formula for calculating cosine similarity is as follows:

[0032]

[0033] The Euclidean distance calculation formula is as follows:

[0034]

[0035] In the formula, γ Similarity (F(x), H(y)) is expressed as cosine similarity; μ Distance (F(x), H(y)) is represented by the Euclidean distance value; F(x) is represented by the feature vector of the current content; H(y) is represented by the feature vector of the historical data; H is represented by the i-th component of the feature vector F(x) of the current extracted content; i (y) is represented as the i-th component in the historical data H(y);

[0036] According to the cosine similarity calculation and Euclidean distance calculation, the content-aware prediction calculation formula is as follows:

[0037]

[0038] In the formula, P(z) represents the dataset z that needs to be accessed; x represents the current content; Y represents the historical data set; Arg max It is expressed as taking the y that maximizes the similarity;

[0039] The data set z is added to the pre-cache area in advance through content-aware prediction. When data is accessed, the data in the pre-cache area is accessed first by default;

[0040] Furthermore, when the entire project data buffer is updated, the corresponding marked data is verified with the feasibility study report according to the cache verification block of the corresponding version in the project data buffer. After the verification is completed, the database is synchronized and the cache data in the project data buffer is eliminated; the project data buffer is initialized, and the data set z is added to the pre-cache area in the project data buffer through content-aware prediction to complete the cache update.

[0041] Furthermore, a feasibility study report automatic generation system based on a large model includes a data acquisition module, a cache control module, a data processing module and an execution module.

[0042] The data collection module includes a project data collection unit and a data cleaning unit; the project data collection unit is used to collect market data, technical data, financial data, economic data, environmental data, risk data, laws and policies, resource data and implementation plans of the project; the data cleaning unit is used to filter out invalid or non-compliant data, detect and delete duplicate records in the data set, integrate data from different sources, and eliminate data conflicts and inconsistencies;

[0043] The cache control module includes a cache construction unit, a cache marking unit and a cache regulation unit; the cache construction unit is used to establish the size and storage structure of the project data buffer according to the project data; the cache marking unit is used to perform corresponding marking processing on different data in the project data buffer; the cache regulation unit is used to set the version of the cache check block and the update of the report content.

[0044] The data processing module includes a template matching generation unit, a data rendering unit, a content perception unit and a data transmission unit; the template matching generation unit is used to dynamically match and generate an adaptive feasibility report template based on the acquired project data information; the data rendering unit is used to render the data in the project data buffer to the corresponding position in the feasibility report template; the content perception unit is used to perform calculations based on the accessed data information and predict the project data that needs to be accessed; the data transmission unit is used to transmit data between various modules and units.

[0045] The execution module includes a report generation unit, a synchronous update unit, a cache execution unit and a database unit; the report generation unit is used to control data interaction to generate a feasibility report; the synchronous update unit is used to update the data information of the feasibility report in real time; the cache execution unit is used to manage the project data buffer; the database unit is used to establish a database table for storage of project data, and at the same time, the data read operations in the project data buffer are synchronously updated to the database table.

[0046] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: by adding cache check blocks and setting cache check blocks of different version levels, the present invention implements fast search, update, deletion and addition operations according to the cache check blocks of the corresponding version levels, and by setting the cache check block version downgrade mode, the operational complexity of cache data is reduced, and the operation speed of the system is greatly improved. The addition of a pre-cache area in the project data buffer and the combination of content perception greatly improve the system performance, reduce response delays, improve cache hit rates, make full use of system resources, improve resource utilization efficiency, and avoid idle resources. By preloading more data into the cache, the frequency of directly reading data from the database is reduced, the burden on the database is reduced, the life of the database is extended, and the overall maintenance cost of the system is reduced. It not only improves the user experience, but also reduces network load and bandwidth costs, creating operating space for the utilization of network resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0048] Figure 1 It is a flow chart of a method for automatically generating a feasibility study report based on a large model according to the present invention;

[0049] Figure 2 It is a structural schematic diagram of a feasibility study report automatic generation system based on a large model of the present invention. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] See also Figure 1 The present invention provides a technical solution: a method for automatically generating a feasibility study report based on a large model. The method for automatically generating a feasibility study report includes the following steps:

[0052] Step S1, establish a project data buffer, obtain data information corresponding to the project, store it in the project data buffer after data cleaning and de-redundancy processing, set a cache check block, the cache check block includes a check block header and a check block footer, add a check block header to the table header according to the data in the project data buffer, add a check block footer to the table footer, and the cache check block follows the non-nested marking rule;

[0053] Step S1-1, establish a project data buffer, and initialize the size and storage structure of the buffer according to the project data;

[0054] Step S1-2, obtaining data information corresponding to the project, and storing it in the project data buffer according to the project data classification after data cleaning and deduplication processing;

[0055] Step S1-3, setting a cache check block, the cache check block includes a check block header for identifying a data table header and a check block tail for identifying a data table tail, and the cache check block follows a non-nested marking rule;

[0056] Step S1-4: Use the cache check block to identify the data in the project data buffer.

[0057] Step S2, set cache verification monitoring to monitor and obtain the changed data in the project data buffer in real time. When the data in the project data buffer is changed, the cache verification monitoring will upgrade the corresponding cache verification block to mark the changed data, and mark the unchanged data with the original version cache verification block;

[0058] Step S2-1, set cache verification monitoring, and monitor and obtain the changed data in the project data buffer in real time;

[0059] Step S2-2, extract the changed data, and upgrade the corresponding cache check block through cache check monitoring, and the upgraded cache check block is used to identify the changed data;

[0060] Step S2-3: The unchanged data is marked with the original version cache check block, following the non-nested marking rule;

[0061] Step S2-4: extract the cache check block data of the version change, update the data in the project data buffer to the feasibility study report, and then update it to the database.

[0062] Step S3, setting a version downgrade block and a downgrade threshold, when the number of data changes in the project data buffer exceeds the downgrade threshold, the version downgrade block downgrades the versions of all cache check blocks;

[0063] Step S4: Analyze and extract the project type and the covered industry information from the project data buffer, and dynamically generate an adaptive personalized feasibility study report template;

[0064] Step S4-1, extract the project type and the covered industry information through the project data buffer analysis, extract the involved industry information from the project data and objectives, determine the industry field covered by the project, set the industry classification as the first-level classification, and set the project data as the second-level classification;

[0065] Step S4-2: Match the adaptation template in the template library according to the project type and industry information of the first-level classification, and adjust the template according to the second-level classification information to generate the final feasibility study report template.

[0066] Step S5: According to the data in the project data buffer and the personalized feasibility study report template, the project data is read out from the project data buffer area, and the project data is matched with the feasibility study report template after data mapping processing to generate a feasibility study report with chapters and contents;

[0067] Step S5-1, loading the personalized feasibility study report template generated by step S4, and reading the project data in the project data buffer area;

[0068] Step S5-2: Render the data to the corresponding position in the feasibility study report template through data matching and data mapping, and generate the corresponding chapters and contents of the feasibility study report.

[0069] Step S6, add a pre-cache area to the project data buffer, obtain the changed project data, predict the project data that needs to be accessed through content perception, and add it to the pre-cache area in advance. When the project data buffer is updated as a whole, repeat the adding method of step S1 to add a cache check block. When the acquired project data information changes, the data in the project data buffer is changed synchronously, and the data in the feasibility study report is updated synchronously through the cache check block version level;

[0070] For example, the project implementation plan data in the feasibility study report is as follows:

[0071] 1. Project implementation plan

[0072] 1.1 Project Progress

[0073] Project period: The total construction period is 2 years and will be implemented in three phases.

[0074] Phase 1: Project launch and preliminary preparation (6 months).

[0075] Phase II: Equipment procurement, installation and commissioning (12 months).

[0076] The third phase: trial operation and formal production (6 months).

[0077] 1.2 Resource Allocation

[0078] Human Resources: The project team includes project managers, technical leaders, procurement leaders, installation and commissioning engineers, etc., totaling 50 people.

[0079] Material resources: The main equipment includes intelligent sensors, data acquisition terminals, servers, cloud computing platforms, etc.

[0080] Change the first phase: project launch and preliminary preparation (6 months) to the first phase: project launch and preliminary preparation (8 months);

[0081] The data is read into the project data buffer and marked by the cache check block. When the cache check monitoring obtains data changes, the changed data "Phase I: Project Startup and Preliminary Preparation (8 months)" is extracted, and the cache check block is upgraded from version 0 to version 1. The cache check block of version 1 is used to mark "Phase I: Project Startup and Preliminary Preparation (8 months)". Other data that have not undergone data operations are marked with the cache check block of version 0. The cache check block of version 1 in the scientific research report is searched, and the data content corresponding to 1 in the project data buffer is synchronously updated to the feasibility study report. Then the data in the project data buffer is written to the database for storage to complete the data update operation.

[0082] Set the version downgrade block and the downgrade threshold, and set the downgrade threshold to δ max , when the number of data changes in the project data buffer exceeds the downgrade threshold δ max When the version downgrade block is used, all cache check blocks are downgraded. The level of each version downgrade is 1. The initial level of the original version cache check block is 0. When the version level of the cache check block is 0, it will no longer be downgraded. Two adjacent cache check blocks of the same version in a database table are merged by default.

[0083] Adding a pre-cache area in the project data buffer, obtaining the changed project data, and predicting the project data to be accessed through content perception, wherein the content perception prediction calculation includes cosine similarity calculation and Euclidean distance calculation;

[0084] The formula for calculating cosine similarity is as follows:

[0085]

[0086] The Euclidean distance calculation formula is as follows:

[0087]

[0088] For example, the current project cost change feature vector and a historical item feature vector The cosine similarity is calculated as follows:

[0089]

[0090] The cosine similarity value is about 0.81. Using similar project data, we can infer the impact of changes in the current project on other related data, thereby guiding the data that needs to be accessed and operated in the next step;

[0091] The Euclidean distance is calculated as follows:

[0092]

[0093] The Euclidean distance is about 3.16, and it is determined that it matches a historical project with a distance of 3.16. The increase in material cost will lead to the following changes:

[0094] Supply chain data: material suppliers, shipping times, supply chain risk assessments, etc.

[0095] Budget data: budget allocation, funding adjustment plans, cost control measures, etc.

[0096] Market price data: material market price trends, market supply conditions, etc.

[0097] For example, α and β are weight coefficients to ensure the comprehensive calculation of the two similarities. Suppose the weight coefficients α = 0.5, β = 0.5, Δ all is the comprehensive similarity score, and the maximum Euclidean distance is set to 10, then the comprehensive similarity score formula is:

[0098]

[0099] According to the comprehensive similarity score Δ all Predict the project data that will be accessed, and obtain the project data with the closest comprehensive similarity score of 0.747. The data that will be accessed after the cost changes are as follows:

[0100] Changes in supply chain data: Delays in delivery from material suppliers, increased transportation costs, and increased supply chain risks;

[0101] Budget data adjustment: budget reallocation, increase of emergency funds, implementation of cost control measures;

[0102] Market price fluctuations: rising material market prices, price adjustments by competitors, and tight market supply;

[0103] Add the above data to the pre-cache area in the project data buffer in advance, so that when the above data is accessed, it can respond quickly;

[0104] In the formula, γ Similarity (F(x), H(y)) is expressed as cosine similarity; μ Distance (F(x), H(y)) is represented by the Euclidean distance value; F(x) is represented by the feature vector of the current content; H(y) is represented by the feature vector of the historical data; H is represented by the i-th component of the feature vector F(x) of the current extracted content; i (y) is represented as the i-th component in the historical data H(y);

[0105] According to the cosine similarity calculation and Euclidean distance calculation, the content-aware prediction calculation formula is as follows:

[0106]

[0107] In the formula, P(z) represents the dataset z that needs to be accessed; x represents the current content; Y represents the historical data set; Arg max It is expressed as taking the y that maximizes the similarity;

[0108] The data set z is added to the pre-cache area in advance through content-aware prediction. When data is accessed, the data in the pre-cache area is accessed by default. When the entire project data buffer is updated, the corresponding marked data is verified with the feasibility study report according to the cache check block of the corresponding version in the project data buffer. After the verification is completed, the database is synchronized and the cache data in the project data buffer is eliminated. The project data buffer is initialized, and the data set z is added to the pre-cache area in the project data buffer through content-aware prediction to complete the cache update.

[0109] See also Figure 2 The present invention provides a technical solution: a feasibility study report automatic generation system based on a large model

[0110] A feasibility study report automatic generation system based on a large model includes a data acquisition module, a cache control module, a data processing module and an execution module;

[0111] The data collection module includes a project data collection unit and a data cleaning unit; the project data collection unit is used to collect market data, technical data, financial data, economic data, environmental data, risk data, laws and policies, resource data and implementation plans of the project; the data cleaning unit is used to filter out invalid or non-compliant data, detect and delete duplicate records in the data set, integrate data from different sources, and eliminate data conflicts and inconsistencies;

[0112] The cache control module includes a cache construction unit, a cache marking unit and a cache regulation unit; the cache construction unit is used to establish the size and storage structure of the project data buffer according to the project data; the cache marking unit is used to perform corresponding marking processing on different data in the project data buffer; the cache regulation unit is used to set the version of the cache check block and the update of the report content.

[0113] The data processing module includes a template matching generation unit, a data rendering unit, a content perception unit and a data transmission unit; the template matching generation unit is used to dynamically match and generate an adaptive feasibility report template based on the acquired project data information; the data rendering unit is used to render the data in the project data buffer to the corresponding position in the feasibility report template; the content perception unit is used to perform calculations based on the accessed data information and predict the project data that needs to be accessed; the data transmission unit is used to transmit data between various modules and units.

[0114] The execution module includes a report generation unit, a synchronous update unit, a cache execution unit and a database unit; the report generation unit is used to control data interaction to generate a feasibility report; the synchronous update unit is used to update the data information of the feasibility report in real time; the cache execution unit is used to manage the project data buffer; the database unit is used to establish a database table for storage of project data, and at the same time, the data read operations in the project data buffer are synchronously updated to the database table.

[0115] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0116] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for automatically generating a feasibility study report based on a large model, characterized in that: The method for automatically generating a feasibility study report comprises the following steps: Step S1, establish a project data buffer, obtain data information corresponding to the project, store it in the project data buffer after data cleaning and de-redundancy processing, set a cache check block, the cache check block includes a check block header and a check block footer, add a check block header to the table header according to the data in the project data buffer, add a check block footer to the table footer, and the cache check block follows the non-nested marking rule; Step S2, set cache verification monitoring to monitor and obtain the changed data in the project data buffer in real time. When the data in the project data buffer is changed, the cache verification monitoring will upgrade the corresponding cache verification block to mark the changed data, and mark the unchanged data with the original version cache verification block; Step S2-1, set cache verification monitoring, and monitor and obtain the changed data in the project data buffer in real time; Step S2-2, extract the changed data, and upgrade the corresponding cache check block through cache check monitoring, and the upgraded cache check block is used to identify the changed data; Step S2-3: The unchanged data is marked with the original version cache check block, following the non-nested marking rule; Step S2-4, extract the cache check block data of the version change, update the data in the project data buffer to the feasibility study report, and then update it to the database; Step S3, setting a version downgrade block and a downgrade threshold, when the number of data changes in the project data buffer exceeds the downgrade threshold, the version downgrade block downgrades the versions of all cache check blocks; Set the version downgrade block and downgrade threshold, and set the downgrade threshold to δ max , when the number of data changes in the project data buffer exceeds the downgrade threshold δ max When the version downgrade block is used, all cache check blocks are downgraded. The level of each version downgrade is 1. The initial level of the original version cache check block is 0. When the version level of the cache check block is 0, it will no longer be downgraded. Two adjacent cache check blocks of the same version in a database table are merged by default. Step S4: Analyze and extract the project type and the covered industry information from the project data buffer, and dynamically generate an adaptive personalized feasibility study report template; Step S5: According to the data in the project data buffer and the personalized feasibility study report template, the project data is read out from the project data buffer area, and the project data is matched with the feasibility study report template after data mapping processing to generate a feasibility study report with chapters and contents; Step S6, add a pre-cache area to the project data buffer, obtain the changed project data, predict the project data that needs to be accessed through content perception, and add it to the pre-cache area in advance. When the project data buffer is updated as a whole, repeat the adding method of step S1 to add a cache check block. When the acquired project data information changes, the data in the project data buffer is changed synchronously, and the data in the feasibility study report is updated synchronously through the cache check block version level.

2. The method for automatically generating a feasibility study report based on a large model according to claim 1 is characterized in that: In step S1, the specific steps of adding a check block header to the table header and adding a check block footer to the table footer according to the data in the project data buffer are as follows: Step S1-1, establish a project data buffer, and initialize the size and storage structure of the buffer according to the project data; Step S1-2, obtaining data information corresponding to the project, and storing it in the project data buffer according to the project data classification after data cleaning and redundancy removal; Step S1-3, setting a cache check block, the cache check block includes a check block header for identifying a data table header and a check block tail for identifying a data table tail, and the cache check block follows a non-nested marking rule; Step S1-4: Use the cache check block to identify the data in the project data buffer.

3. The method for automatically generating a feasibility study report based on a large model according to claim 2 is characterized in that: In step S4, the specific steps of dynamically generating an adapted personalized feasibility study report template are as follows: Step S4-1, extract the project type and the covered industry information through the project data buffer analysis, extract the involved industry information from the project data and objectives, determine the industry field covered by the project, set the industry classification as the first-level classification, and set the project data as the second-level classification; Step S4-2: Match the adaptation template in the template library according to the project type and industry information of the first-level classification, and adjust the template according to the second-level classification information to generate the final feasibility study report template.

4. The method for automatically generating a feasibility study report based on a large model according to claim 3 is characterized by: In step S5, the specific steps of matching and generating a feasibility study report with chapters and contents are as follows: Step S5-1, loading the personalized feasibility study report template generated by step S4, and reading the project data in the project data buffer area; Step S5-2: Render the data to the corresponding position in the feasibility study report template through data matching and data mapping, and generate the corresponding chapters and contents of the feasibility study report.

5. The method for automatically generating a feasibility study report based on a large model according to claim 4 is characterized in that: In step S6, a pre-cache area is added to the project data buffer, the changed project data is obtained, and the project data to be accessed is predicted by content perception, wherein the content perception prediction calculation includes cosine similarity calculation and Euclidean distance calculation; The formula for calculating cosine similarity is as follows: The Euclidean distance calculation formula is as follows: In the formula, γ Similarity (F(x), H(y)) is expressed as cosine similarity; μ Distance (F(x), H(y)) is represented by the Euclidean distance value; F(x) is represented by the feature vector of the current content; H(y) is represented by the feature vector of the historical data; F i (x) is represented as the i-th component in the current extracted content feature vector F(x); H i (y) is represented as the i-th component in the historical data H(y); According to the cosine similarity calculation and Euclidean distance calculation, the content-aware prediction calculation formula is as follows: In the formula, P(z) represents the dataset z that needs to be accessed; x represents the current content; Y represents the historical data set; Arg max It is expressed as taking the y that maximizes the similarity; The data set z is added to the pre-cache area in advance through content-aware prediction. When data is accessed, the data in the pre-cache area is accessed first by default; When updating the entire project data buffer, the corresponding marked data is verified with the feasibility study report according to the cache verification block of the corresponding version in the project data buffer. After the verification is completed, the database is synchronized and the cache data in the project data buffer is eliminated; the project data buffer is initialized, and the data set z is added to the pre-cache area in the project data buffer through content-aware prediction to complete the cache update.

6. A system for automatically generating a feasibility study report based on a large model, applied to a method for automatically generating a feasibility study report based on a large model as claimed in any one of claims 1 to 5, characterized in that: The feasibility study report automatic generation system includes a data acquisition module and a cache control module; The data collection module includes a project data collection unit and a data cleaning unit; the project data collection unit is used to collect market data, technical data, financial data, economic data, environmental data, risk data, laws and policies, resource data and implementation plans of the project; the data cleaning unit is used to filter out invalid or non-compliant data, detect and delete duplicate records in the data set, integrate data from different sources, and eliminate data conflicts and inconsistencies; The cache control module includes a cache construction unit, a cache marking unit and a cache regulation unit; the cache construction unit is used to establish the size and storage structure of the project data buffer according to the project data; the cache marking unit is used to perform corresponding marking processing on different data in the project data buffer; the cache regulation unit is used to set the version of the cache check block and the update of the report content.

7. The system for automatically generating a feasibility study report based on a large model according to claim 6 is characterized by: The feasibility study report automatic generation system also includes a data processing module; The data processing module includes a template matching generation unit, a data rendering unit, a content perception unit and a data transmission unit; the template matching generation unit is used to dynamically match and generate an adapted feasibility study report template according to the acquired project data information; The data rendering unit is used to render the data in the project data buffer to the corresponding position in the feasibility study report template; the content perception unit is used to calculate according to the accessed data information and predict the project data that needs to be accessed; the data transmission unit is used to transmit data between various modules and units.

8. The system for automatically generating a feasibility study report based on a large model according to claim 6 is characterized by: The feasibility study report automatic generation system also includes an execution module; The execution module includes a report generation unit, a synchronous update unit, a cache execution unit and a database unit; the report generation unit is used to control data interaction to generate a feasibility study report; the synchronous update unit is used to update the data information of the feasibility study report in real time; The cache execution unit is used to manage the project data buffer; the database unit is used to establish a database table for storing the project data, and at the same time, the data read operation in the project data buffer is synchronously updated to the database table.

Images