Intelligent evaluation project value comprehensive index management method

Abstract

The invention discloses an intelligent evaluation project value comprehensive index management method, and belongs to the field of data processing, and the method comprises the steps: building a project value evaluation model, and carrying out the training; making a data collection template, obtaining various data of each energy project, classifying and establishing an energy project basic database; the collected data samples are preprocessed, key indexes are preset, the data are screened and cleaned, if the data meet the requirements, the next step is executed, and if not, declaration is returned; evaluating each energy project to obtain an evaluation result; the project value evaluation model is optimized according to the evaluation result; and based on scores of the energy projects evaluated by the optimized project value evaluation model, pushing the scores to a project reserve library according to types in a classified manner, and preferentially distributing declaration indexes by a project declaration system according to projects with high scores and low scores until the declaration indexes are distributed completely. According to the invention, the big data analysis capability of the AI technology is fully utilized, and the transformation of the review process from digitization to intelligentization is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a management method for intelligently judging comprehensive indicators of project value. Background Art

[0002] The review mechanism of the traditional project application system has long been criticized for being too subjective. The personal preferences and subjective judgments of the review experts often become the key factors affecting the application results, which to a certain extent weakens the objectivity and fairness of the review. In addition, the ambiguity of the review standards and processes is like a fog shrouding the minds of applicants, making it difficult for applicants to see the full picture and exact results of the review, which invisibly exacerbates the uncertainty of the application process and the anxiety of applicants.

[0003] With the rapid development of AI technology in recent years, its deep integration and outstanding performance in multiple fields are driving the transformation and upgrading of all walks of life with unprecedented power. AI not only demonstrates its powerful data processing and analysis capabilities, but also provides the possibility to solve the shortcomings of the traditional review system with its intelligent and efficient characteristics.

[0004] Therefore, it is necessary to develop a management method for intelligently judging the comprehensive indicators of project value, and use AI technology to enable the project application system to overcome its existing drawbacks and solve the problems existing in the traditional project application system. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide a management method for intelligently judging the comprehensive indicators of project value, making full use of the big data analysis capabilities of AI technology to realize the transformation of the review process from digitalization to intelligence; the project application system can automatically and objectively conduct multi-dimensional and in-depth analysis and evaluation of the application projects, effectively eliminating the interference of human factors on the analysis results, and ensuring the fairness and objectivity of the review; at the same time, the project application system also has the ability of continuous self-learning and optimization, can continuously adapt to changing needs, and improve the accuracy and efficiency of the review.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] A management method for intelligently evaluating comprehensive indicators of project value, including:

[0008] A large amount of historical project data in the project application system is used as a training set to build a project value assessment model. Intelligent AI uses a recurrent neural network architecture to train the project value assessment model.

[0009] Create data collection templates, obtain data on various energy projects, classify them according to project application indicators, and establish a basic database for energy projects;

[0010] The collected data samples are pre-processed and key indicators are preset. The project application system screens and cleans the data according to key factors. If the key indicators are matched, the application process can be entered into the next step. If not, the project application will be automatically returned. All energy projects that enter the next step of the application process constitute the project reserve pool.

[0011] Use the trained project value assessment model to assess each energy project in the project reserve pool and obtain the assessment results;

[0012] The project value assessment model is then optimized based on the assessment results of each energy project;

[0013] The scores of each energy project evaluated based on the optimized project value assessment model are pushed to the project reserve library by type. The project application system allocates application indicators based on the project scores, with projects with high scores given priority until the application indicators are fully allocated.

[0014] A further improvement of the technical solution of the present invention is that the data include at least: total project investment, annual investment plan, project type, energy type, new construction / continuation type, planned power generation, supporting energy storage power, supporting energy storage capacity, supporting hydrogen production, project location, land area, planned construction period, supporting peak-shaving measures, supporting peak-shaving measures, planned grid access time, planned total grid connection time, construction content, shareholder information, expected investment recovery period, expected rate of return on investment, status of preliminary procedures, area of ​​land expropriated and leased, area of ​​land implemented, and whether the interests of the masses are implemented.

[0015] A further improvement of the technical solution of the present invention is that the preliminary procedures at least include: whether the land lease procedures have been obtained, land pre-examination documents, environmental impact assessment report, soil and water conservation plan, feasibility study report on forest and grassland use, and land comprehensive utilization plan.

[0016] A further improvement of the technical solution of the present invention is that the preset key indicators include at least: the total investment of the project is not less than 100 million yuan, the capacity of the generator set is between 300,000 kW and 2,000,000 kW, and the planned construction period is not more than 36 months.

[0017] A further improvement of the technical solution of the present invention is that when the project value assessment model evaluates each energy project, the project application system can assign initial weights to the set indicator values, and the sum of the weights of the indicator values ​​is 1. The project application system calculates the score S for each indicator value according to the set scoring standard, which are S1, S2, S3...Sn respectively. After all calculations are completed, the total score M of each project is calculated in combination with the weight:

[0018]

[0019] Where Si is the value of each indicator; Wi is the weight value.

[0020] The further improvement of the technical solution of the present invention is that: by comprehensively considering various evaluation indicators, the project application system can automatically generate a comprehensive and objective comprehensive score, providing a strong basis for project evaluation;

[0021] For the "more serious" deduction items identified during the comparison process, the project application system will intelligently generate targeted optimization suggestions, which can help the project team clarify the direction of improvement and further improve the project quality and competitiveness.

[0022] A further improvement of the technical solution of the present invention is that when the project value assessment model is tuned, it at least includes adjusting model parameters and replacing model architecture; it can also increase the offset to strengthen the logic and improve the data value of the reference results.

[0023] The further improvement of the technical solution of the present invention is that: the offset setting is divided into two types: elastic offset and vector offset;

[0024] Elastic offset stretches or squeezes the width of the range. The algorithm is as follows:

[0025] Q min =S min +(1-V 弹性 )(S max -S min ) / 2

[0026] Q max =Q max -(1-V 弹性 )(S max -S min ) / 2

[0027] Vector offset is a linear translation of the range domain. The algorithm is as follows:

[0028] R min =S min +V 向量

[0029] R max =S max +V 向量

[0030] In the formula, S max Indicates the maximum value of the data before the offset; S min Indicates the minimum value of the data before the offset; V 弹性 Indicates the set elastic offset; Q max Indicates the maximum value of the data after elastic offset; Q min Represents the minimum value of the data after elastic offset; Rmax Indicates the maximum value of the data after vector offset; R min Indicates the minimum value of the data after the vector is offset; V 向量 Indicates the set vector offset;

[0031] Elastic offset and vector offset are compensations for the limited amount of initial training sample data and the bias in the project value assessment model of the training results. As new data samples continue to flow in, the project value assessment model trained by intelligent AI will become more accurate, the offset will be gradually reduced, and eventually a mature project value assessment model with 0 offset will be achieved.

[0032] A further improvement of the technical solution of the present invention is that the scores of various energy projects evaluated based on the optimized project value assessment model are pushed to the project reserve library by type classification, and the project application system allocates application indicators to projects with high scores first, until the application indicators are fully allocated. This process is tightly integrated into the approval process engine of the project application system, ensuring a smooth and efficient approval process.

[0033] A further improvement of the technical solution of the present invention is that all processed data are saved as historical records to facilitate subsequent data output and visual display, and can also provide sample resources for the continuous optimization and enhanced training of the project value assessment model.

[0034] Due to the adoption of the above technical solution, the technical progress achieved by the present invention is:

[0035] 1. The present invention makes full use of the big data analysis capabilities of AI technology to realize the transformation of the review process from digitalization to intelligence; the project application system can automatically and objectively conduct multi-dimensional and in-depth analysis and evaluation of the application projects, effectively eliminating the interference of human factors on the analysis results, and ensuring the fairness and objectivity of the review; at the same time, the project application system also has the ability of continuous self-learning and optimization, can continuously adapt to changing needs, and improve the accuracy and efficiency of the review.

[0036] 2. The project application system in the present invention can generate evaluation reports in real time and provide optimization suggestions, helping the project team to quickly locate problems and take effective measures to improve and explore best practices.

[0037] 3. The project application system in the present invention starts with the training of the project value assessment model, to project information reporting, assessment, business flow approval, data output and display, and the final result becomes a data sample to participate in the model training again, forming a complete closed loop, autonomous iteration, continuous improvement, and a comprehensive business chain.

[0038] 4. The project application system in the present invention is a means of assisting decision-making, which can objectively and impartially evaluate the value of the project and eliminate or reduce the impact of "artificial" subjective ideas on project construction decisions.

[0039] 5. The project application system of the present invention supports enhanced training based on new sample data, can continuously adapt to market changes and technological development, and maintain the advancement and accuracy of the project value assessment model. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

[0041] Figure 1 It is a flow chart of a management method for intelligently judging comprehensive indicators of project value provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0042] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or apparatus.

[0043] The present invention provides a management method for intelligently judging the comprehensive indicators of project value. It is a method for comprehensively evaluating the construction value of a project using a deep learning algorithm. A large amount of historical project data is used as a training set to construct a project value evaluation model. During the training process of the project value evaluation model, focus is placed on multiple dimensional data of the project, including but not limited to market demand, technical feasibility, resource investment, expected returns, etc. Through multiple rounds of iterative training, the parameters of the project value evaluation model are continuously optimized to improve the accuracy and generalization ability of the evaluation.

[0044] Based on the training of the project value assessment model, the project application system can automatically identify the key features that affect the project value and set reasonable thresholds based on statistical analysis and expert knowledge. These key features and their thresholds will serve as an important basis for subsequent project evaluation to ensure the objectivity and fairness of the evaluation results.

[0045] Combined with advanced technology applications such as approval processes, report printing, and data analysis visualization, it can greatly improve management standardization and work efficiency.

[0046] This intelligent management method for judging the comprehensive indicators of project value is a means of assisting decision-making. It can objectively and impartially evaluate the value of projects and eliminate or reduce the impact of "human" subjective ideas on project construction decisions. It has a wide range of application scenarios and is suitable for project investment, government agencies, scientific research institutes and other project review fields. It has considerable guiding significance for investment decisions and evaluation approvals for large-scale project construction such as power facilities, roads and bridges, water conservancy projects, and scientific research, improving project success rates, optimizing project construction processes, and improving management levels.

[0047] The present invention is further described in detail below with reference to the accompanying drawings and embodiments:

[0048] like Figure 1 As shown, a management method for intelligently judging the comprehensive indicators of project value includes the following steps:

[0049] Step 1: Use a large amount of historical project data in the project application system as a training set to build a project value assessment model. Intelligent AI uses a recurrent neural network architecture to train the project value assessment model.

[0050] Step 2, make a data collection template to obtain various data of each energy project, such as: total project investment, annual investment plan, project type, energy type, new / continued type, planned power generation, supporting energy storage power, supporting energy storage capacity, supporting hydrogen production, project location, land area, planned construction period, supporting peak-shaving measures, supporting peak-shaving measures, planned grid access time, planned full grid connection time, construction content, shareholder information, expected investment recovery period, expected investment return rate, preliminary procedures (such as: whether land lease procedures have been obtained, land pre-examination documents, environmental impact assessment report, soil and water conservation plan, feasibility study report on forest and grassland use, land comprehensive utilization plan, etc.), area of ​​land expropriated, land area implemented, whether the interests of the masses have been implemented, etc., and classify according to project application indicators to establish a basic database of energy projects.

[0051] Step 3, pre-process the collected data samples and preset key indicators, such as: the total investment of the project is not less than 100 million yuan, the capacity of the generator set is between 300,000 kW and 2,000,000 kW, and the planned construction period is not more than 36 months. The project application system screens and cleans the data according to key factors. If the key indicators are matched, the next application process can be entered. If not matched, the project application will be automatically returned; all energy projects that enter the next application process constitute a project reserve library; this can exclude data with missing values, outliers, duplicate values, etc., to improve the quality of sample data and reduce interference during model training.

[0052] Step 4: Use the trained project value assessment model to assess each energy project in the project reserve pool to obtain an assessment result;

[0053] The sum of the weights of each indicator value is 1. The project application system calculates the score S for each indicator value according to the set scoring standard, which are S1, S2, S3...Sn. After all are completed, the total score M of each project is calculated based on the weight, as shown in the following formula:

[0054]

[0055] Where Si is the value of each indicator; Wi is the weight value.

[0056] In the project application system, initial weights are assigned to the set indicator values. The initial weight configuration is shown in Table 1 below:

[0057] Table 1 Detailed table of initial weight configuration of various indicator values

[0058]

[0059]

[0060] This process is closely based on the project value assessment model constructed in step 1. By comprehensively considering various assessment indicators, the project application system can automatically generate a comprehensive and objective comprehensive score, providing a strong basis for project review. More importantly, for the "more serious" deduction items identified during the comparison process, the project application system will intelligently generate targeted optimization suggestions to help the project team clarify the direction of improvement and further improve the quality and competitiveness of the project. This function not only reflects the intelligence and practicality of the project application system.

[0061] Step 5: Optimize the project value assessment model based on the assessment results of each energy project;

[0062] When optimizing the project value assessment model, at least the model parameters and model architecture must be adjusted. You can also increase the offset to strengthen the logic and improve the data value of the reference results. There are two types of result offset settings: elastic offset and vector offset. Elastic offset mainly stretches or squeezes the width of the range, while vector offset linearly translates the range.

[0063] Elastic offset stretches or squeezes the width of the range. The algorithm is as follows:

[0064] Q min =S min +(1-V 弹性 )(S max -S min ) / 2

[0065] Q max =Q max -(1-V 弹性 )(S max -S min ) / 2

[0066] Vector offset is a linear translation of the range domain. The algorithm is as follows:

[0067] R min =S min +V 向量

[0068] R max =S max +V 向量

[0069] In the formula, S max Indicates the maximum value of the data before the offset; S min Indicates the minimum value of the data before the offset; V 弹性 Indicates the set elastic offset; Q max Indicates the maximum value of the data after elastic offset; Q min Represents the minimum value of the data after elastic offset; R max Indicates the maximum value of the data after vector offset; R min Indicates the minimum value of the data after the vector is offset; V 向量 Indicates the set vector offset;

[0070] Elastic offset and vector offset are compensations for the limited amount of initial training sample data and the bias in the project value assessment model of the training results. As new data samples continue to flow in, the project value assessment model trained by intelligent AI will become more accurate, the offset will be gradually reduced, and eventually a mature project value assessment model with 0 offset will be achieved.

[0071] Step 6: Based on the scores of each energy project evaluated by the optimized project value assessment model, they are pushed to the project reserve library by type classification. The project application system allocates application indicators to projects with high scores first, until the application indicators are fully allocated.

[0072] This process is tightly integrated into the approval process engine of this management system, ensuring a smooth and efficient approval process. All processed data is carefully saved as historical records, which not only facilitates subsequent data output and visual display, but also provides valuable sample resources for the continuous optimization and enhanced training of project value assessment models. Through these continuously accumulated and updated data assets, the intelligence and accuracy of the project application system can be continuously improved, providing more solid data support for decision-making.

[0073] The contents not described in detail in this specification belong to the prior art known to professional and technical personnel in this field.

[0074] In summary, the present invention provides a management method for intelligently judging the comprehensive indicators of project value, which makes full use of the big data analysis capabilities of AI technology and realizes the transformation of the review process from digital to intelligent. The project application system can automatically and objectively conduct multi-dimensional and in-depth analysis and evaluation of the application projects, effectively eliminating the interference of human factors on the analysis results, and ensuring the fairness and objectivity of the review. At the same time, the system also has the ability of continuous self-learning and optimization, can continuously adapt to changing needs, and improve the accuracy and efficiency of the review.

[0075] The use of this invention in the project application system is not only an innovation of the traditional review mechanism, but also a vivid practice of AI technology in promoting social equity and efficiency improvement. It will provide a more transparent and fair review environment for project applicants, help more high-quality projects stand out, and promote the sustainable and healthy development of the social economy.

[0076] Finally, 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A management method for intelligently evaluating comprehensive indicators of project value. It is characterized in that include: A large amount of historical project data in the project application system is used as a training set to build a project value assessment model. Intelligent AI uses a recurrent neural network architecture to train the project value assessment model. Create data collection templates, obtain data on various energy projects, classify them according to project application indicators, and establish a basic database for energy projects; The collected data samples are pre-processed and key indicators are preset. The project application system screens and cleans the data according to key factors. If the key indicators are matched, the application process can be entered into the next step. If not, the project application will be automatically returned. All energy projects that enter the next step of the application process constitute the project reserve pool. Use the trained project value assessment model to assess each energy project in the project reserve pool and obtain the assessment results; The project value assessment model is then optimized based on the assessment results of each energy project; The scores of each energy project evaluated based on the optimized project value assessment model are pushed to the project reserve library by type. The project application system allocates application indicators based on the project scores, with projects with high scores given priority until the application indicators are fully allocated.

2. According to the management method of a comprehensive index of intelligent evaluation of project value according to claim 1, Features: The data mentioned include at least: total project investment, annual investment plan, project type, energy type, new / continued construction type, planned power generation, supporting energy storage power, supporting energy storage capacity, supporting hydrogen production, project location, land area, planned construction period, supporting peak-shaving measures, supporting peak-shaving measures, planned grid access time, planned total grid connection time, construction content, shareholder information, expected investment recovery period, expected rate of return on investment, status of preliminary procedures, area of ​​land expropriated and leased, area of ​​land implemented, and whether the interests of the masses are implemented.

3. According to the management method of a comprehensive index of intelligent evaluation of project value according to claim 2, Features: The status of the preliminary procedures shall at least include: whether the land lease procedures have been obtained, land pre-examination documents, environmental impact assessment report, soil and water conservation plan, feasibility study report on forest and grassland use, and comprehensive land utilization plan.

4. According to the management method of a comprehensive index of intelligent evaluation of project value according to claim 1, Features: The preset key indicators include at least: the total project investment is not less than 100 million yuan, the capacity of the generator set is between 300,000 kW and 2,000,000 kW, and the planned construction period is not more than 36 months.

5. According to the management method of the comprehensive index of intelligent evaluation of project value according to claim 1, Features: When the project value assessment model evaluates each energy project, the project application system can assign initial weights to the set indicator values. The sum of the weights of each indicator value is 1. The project application system calculates the score S for each indicator value according to the set scoring standard, which are S1, S2, S3...Sn. After all calculations are completed, the total score M of each project is calculated based on the weight: Where Si is the value of each indicator; Wi is the weight value.

6. According to the management method of a comprehensive index of intelligent evaluation of project value according to claim 1, Features: By comprehensively considering various evaluation indicators, the project application system can automatically generate a comprehensive and objective comprehensive score, providing a strong basis for project evaluation; For the "more serious" deduction items identified during the comparison process, the project application system will intelligently generate targeted optimization suggestions, which can help the project team identify the direction of improvement and further enhance the project quality and competitiveness.

7. According to the management method of the comprehensive index of intelligent evaluation of project value according to claim 1, Features: When tuning the project value assessment model, it at least includes adjusting model parameters and changing model architecture; it can also add offsets to strengthen the logic and improve the data value of the reference results.

8. According to the management method of the comprehensive index of intelligent evaluation of project value according to claim 7, Features: There are two types of offset settings: elastic offset and vector offset; Elastic offset stretches or squeezes the width of the range. The algorithm is as follows: Q min =S min +(1-V 弹性 )(S max -S min ) / 2 Q max =Q max -(1-V 弹性 )(S max -S min ) / 2 Vector offset is a linear translation of the range domain. The algorithm is as follows: R min =S min +V 向量 R max =S max +V 向量 In the formula, S max Indicates the maximum value of the data before the offset; S min Indicates the minimum value of the data before the offset; V 弹性 Indicates the set elastic offset; Q max Indicates the maximum value of the data after elastic offset; Q min Indicates the minimum value of the data after elastic offset; R max Indicates the maximum value of the data after vector offset; R min Indicates the minimum value of the data after the vector is offset; V 向量 Indicates the set vector offset; Elastic offset and vector offset are compensations for the limited amount of initial training sample data and the bias in the project value assessment model of the training results. As new data samples continue to flow in, the project value assessment model trained by intelligent AI will become more accurate, the offset will be gradually reduced, and eventually a mature project value assessment model with 0 offset will be achieved.

9. According to the management method of the comprehensive index of intelligent evaluation of project value according to claim 1, Features: The scores of each energy project evaluated based on the optimized project value assessment model are pushed to the project reserve library by type. The project application system gives priority to the application indicators of projects with high scores according to the project scores until the application indicators are fully allocated. This process is tightly integrated into the approval process engine of the project application system, ensuring a smooth and efficient approval process.

10. According to the management method of the comprehensive index of intelligent evaluation of project value according to claim 1, Features: All processed data are saved as historical records to facilitate subsequent data output and visualization, and can also provide sample resources for the continuous optimization and enhanced training of project value assessment models.

Images