A method and system for identifying collaborative relationship features in power engineering projects

Abstract

The present invention discloses a method and system for identifying the characteristics of collaborative relationships of power engineering projects. The method and system are based on the Apriori data mining association algorithm to identify and mine the characteristics of collaborative relationships among investment plans, milestone plans and funding plans in the entire process of power engineering projects. By defining three calculation rules of support, confidence and lift, an empirical analysis is performed based on historical sample data of engineering projects in the past five years, and reasonable thresholds are set for support, confidence and lift to find out the implicit cross-reference relationship among the three plans of investment plan, milestone plan and funding plan in the engineering projects. The management and control strategies are summarized from the association rule results to provide an association relationship basis for process risk identification during project implementation, thereby achieving the purpose of enhancing the coordination degree of the entire project process.

Description

Technical Field

[0001] The present invention relates to the field of machine learning and big data processing in the power grid industry, and more specifically, to a method and system for identifying collaborative relationship features of power engineering projects. Background Art

[0002] The investment plan system of power engineering projects consists of investment plan, capital plan and milestone plan. At present, the management of the three plans lacks cross-reference and connection, forming an isolated plan. There are three main problems: First, the source of the investment plan is not clear enough. At present, the investment is basically determined by the total investment of the project according to its own investment capacity, and it has not been well based on its own needs to obtain scientific results; second, the scope of the milestone plan is small, with only new construction milestones and production milestones. For the entire cycle of the project, the coverage is not enough, and there is no good guidance for the investment plan and capital plan; third, the capital plan is not taken seriously enough, and the preparation of the capital plan is basically equivalent to the investment plan, and the role of the capital plan as a pre-control of capital payment has not been reflected. Therefore, there is an urgent need for a technical solution that can present the cross-reference relationship between the three plans of investment, capital and milestone, and by building a relationship model, the three plans can be combined into an integrated plan system, and improved in practical applications to achieve more scientific investment plan management.

[0003] The prior art discloses a corporate financial diagnosis system based on big data, including an intelligent collection module for intelligently collecting relevant corporate information, conducting data analysis based on big data, and forming an enterprise data set; also including a verification module for calling the enterprise data set, cross-checking the financial statements of the enterprise, and checking whether the cross-check relationship within and between tables is normal; if there is an abnormality, the cross-check relationship abnormal point is output; if the cross-check relationship is seriously abnormal, the financial analysis is stopped, and the cross-check result is output; if the cross-check relationship warns of non-serious abnormalities, the financial analysis is continued; also including an evaluation unit for identifying and evaluating non-serious abnormal enterprises, and conducting comprehensive financial analysis, and outputting fraud index and health index. Although this solution can identify the cross-check relationship, it cannot realize the integration of investment plan, capital plan and milestone plan in the field of power engineering. Summary of the invention

[0004] The primary purpose of the present invention is to provide a method for identifying collaborative relationship features of power engineering projects. By cleaning multi-source heterogeneous data, a relationship feature recognition and project management knowledge graph is established to provide power companies with full-process monitoring of engineering project implementation, as well as identification and monitoring services based on the collaborative relationship among their investment plans, milestone plans and funding plans.

[0005] A further object of the present invention is to provide a system for identifying collaborative relationship characteristics of power engineering projects.

[0006] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0007] A method for identifying collaborative relationship characteristics of power engineering projects, comprising the following steps:

[0008] S1: Collect information on power projects that have been included in investment plans, implemented, or generated funding payments in the past five years across departments;

[0009] S2: Determine whether the investment plan, milestone plan and funding plan of the collected projects match. If they match, proceed to step S4; if not, proceed to step S3;

[0010] S3: Build a three-plan matching relationship feature index library and output the three-plan matching relationship situation;

[0011] S4: Build a characteristic indicator database of the synergistic relationship between the three plans and output the synergistic relationship between the three plans;

[0012] S5: Integrate the three-plan matching relationship characteristic index library and the three-plan collaborative relationship characteristic index library to output the results.

[0013] Preferably, the project information in step S1 includes basic information and process-related data, wherein the basic information includes the project number, project type and implementation unit, and the process-related data includes the node time, investment and payment amount in the project implementation.

[0014] Preferably, the step S2 is preceded by the following steps:

[0015] Among the projects collected in step S1, unfinished projects and projects with missing information data are eliminated.

[0016] Preferably, in step S2, it is determined whether the investment plan, milestone plan and funding plan of the collected projects match, specifically:

[0017] Using the project number as a unique identifier, summarize the investment plans, milestone plans and funding plans of the past five years, and batch search whether each project exists in the investment plan, milestone plan and funding plan at the same time. If not, the project is a project with management problems, and the three plans do not match, and go to step S3; if so, the three plans of the project match, and go to step S4.

[0018] Preferably, in step S3, a three-plan matching relationship characteristic index library is constructed, and the three-plan matching relationship is output, specifically:

[0019] S3.1: Calculation of the three-plan matching relationship index library: Combined with the project full process control stage, through system sorting and historical experience, sort out all relevant factors that affect the three-plan matching and cause management problem projects in the full process control of engineering projects, and classify these relevant factors into quantifiable and calculable indicators according to the management problem projects to form an index library. The indicators are screened through the expert scoring method, and then the indicators of each project in the unmatched project library are calculated to form a database;

[0020] S3.2: Indicator matching statistics: Through the association rule set, the implicit and valuable connections between the indicator libraries are mined. The project refers to the various attributes of any non-empty record, and the item set refers to the set of all attributes. All the attributes of all non-empty records are stored in the database D). Assume that there are N project data in the sample space Ω, and each project contains the deviation of M key indicators. The deviation of each indicator in different projects is represented by 0 or 1. The deviation is 1 and the non-deviation is 0. The management rules of the key control indicators of the project can be expressed in three forms: support, confidence and lift. The support refers to the probability that the antecedent X and the consequent Y of the rule in the item set occur at the same time, indicating the probability that the control indicators x and y of the project sample data deviate at the same time. σ(X,Y) represents the total number of times x and y deviate at the same time. N is the number of project samples. The support Support(X→Y) is calculated as follows:

[0021]

[0022]

[0023] Confidence refers to the probability of the occurrence of the rule's subsequent term Y under the premise that the rule's antecedent term X occurs in the term set. It means the probability that the control indicator y will deviate when the control indicator x deviates. The calculation method of confidence (X→Y) is as follows:

[0024]

[0025] Lift refers to the ratio of the probability of the rule-based item Y occurring after the preceding item X in the item set to the probability of the rule-based item Y occurring. It indicates the frequency with which the occurrence of item set X leads to the occurrence of item set Y. The calculation method of lift (X→Y) is as follows:

[0026]

[0027] S3.3: Eliminate matching relationship conclusions that do not meet most of the indicators: Based on the preliminary data calculation results, the thresholds of support, confidence, and lift are set to 0.08, 0.4, and 1, respectively. Association rules that meet the thresholds indicate that the antecedent and consequent of the rules also have a certain positive correlation. Association rules that are lower than the thresholds indicate that the correlation between the antecedent and consequent of the rules is weak, so such association rules are eliminated;

[0028] S3.4: Determination of the characteristic index library of the matching relationship between the three plans: According to the required association rule scale and traceability requirements, adjust the optimization threshold, determine the indicators corresponding to the strongly related association rules, and form the characteristic index library of the matching relationship between the three plans;

[0029] S3.5: Output of the matching relationship between the three plans of the project: Select association rules that meet the support, confidence and lift thresholds, interpret the relationship between the antecedents and the subsequent items of the association rules, and output the matching relationship between the three plans.

[0030] Preferably, in step S3.2, the direct correlation relationship in the project matching index is analyzed by using the Apriori algorithm to mine frequent item sets of project data.

[0031] Preferably, the step S3.2 follows the core idea of ​​multiple rounds of iterations to gradually mine frequent itemsets, specifically:

[0032] In the first round of iteration, the minimum support threshold is determined and compared, and the candidate set C is selected. 1 Generate frequent item sets L 1 ;

[0033] In subsequent multiple loop iterations, candidate C is generated from the frequent item set index set. k+1 The indicator set is obtained through the frequent item set L k The indicator set is connected to generate candidate C k+1 At the same time, relying on the two properties that any non-empty subset of the frequent indicator set is frequent and any superset of the infrequent indicator set is infrequent, the candidate C is completed by deleting the infrequent item sets in the candidate indicator set. k+1 Pruning of indicator sets;

[0034] The iterative process continues until no new frequent indicator sets can be generated. Then the confidence and lift of all frequent multi-item sets are calculated respectively, and it is determined whether they exceed the threshold. The algorithm outputs all frequent multi-item association rules that meet the requirements.

[0035] Preferably, in step S3.2, the deviation of each indicator of each project in the mismatched project library is collected by power engineering related management experts through a survey questionnaire method, and statistics are performed. If the management expert believes that a certain indicator deviation causes the project mismatch, it is marked as 1, otherwise it is marked as 0, and the association relationship is mined through an R language program.

[0036] Preferably, in step S4, a three-plan collaborative relationship characteristic index library is constructed to output the three-plan collaborative relationship situation, specifically:

[0037] S4.1: Calculation of the three-plan coordination relationship indicator database. For matching projects that exist in the investment plan, milestone plan, and funding plan at the same time, study whether the corresponding investment completion progress, engineering implementation progress, and funding payment progress during the entire implementation process of the project are coordinated. Therefore, based on the available milestone data, the project full-process control stage is divided into three stages: construction, settlement, and final settlement. Through system sorting, historical experience, and expert scoring methods, sort out all possible relevant factors that affect the coordination of the three plans in the full-process control of engineering projects, form quantifiable and calculable indicators, classify various indicators according to the control stage, form an indicator database, screen the indicators through the expert scoring method, and then calculate the indicators of each project in the project database to form a database;

[0038] S4.2: Indicator matching statistics: Apriori algorithm is used to mine frequent item sets of project data. Through questionnaire survey, the deviation of each indicator of each project in the project library is collected by power engineering management experts to conduct statistics. If the manager believes that a certain indicator deviation causes the project to be mismatched, it is marked as 1, otherwise it is marked as 0. The association relationship is mined through R language program;

[0039] S4.3: Eliminate matching relationship conclusions that do not meet most of the indicators: Based on the preliminary data calculation results, the thresholds of support, confidence, and lift are set to 0.06, 0.3, and 1, respectively. Association rules that meet the thresholds indicate that the antecedent and consequent of the rules also have a certain positive correlation. Association rules that are lower than the thresholds indicate that the correlation between the antecedent and consequent of the rules is weak, so such association rules are eliminated;

[0040] S4.4: Determination of the characteristic index library of the synergistic relationship between the three plans: According to the required association rule scale and traceability requirements, adjust the optimization threshold, determine the indicators corresponding to the strongly related association rules, and form the characteristic index library of the synergistic relationship between the three plans;

[0041] S4.5: Output of the synergistic relationship of the three plans of the project: Select association rules that meet the support, confidence and lift thresholds, interpret the relationship between the antecedent and subsequent items of the association rules, and output the synergistic relationship of the three plans.

[0042] A power engineering project collaborative relationship feature identification system, the feature identification system is applied to the above-mentioned power engineering project collaborative relationship feature identification method, the system comprises:

[0043] A collection module, which is used to collect information on projects in power engineering projects that have been included in investment plans, implemented, or generated funding payments in the past five years across departments;

[0044] A matching module, which is used to determine whether the investment plan, milestone plan and funding plan in the collected projects match. If they match, the module enters the matching relationship feature module; if not, the module enters the collaborative relationship feature module;

[0045] A matching relationship feature module, which is used to construct a three-plan matching relationship feature index library and output the three-plan matching relationship status;

[0046] A collaborative relationship feature module, which is used to construct a three-plan collaborative relationship feature index library and output the three-plan collaborative relationship status;

[0047] The output module is used to integrate the three-plan matching relationship feature index library and the three-plan collaborative relationship feature index library to output the results.

[0048] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0049] 1. The present invention integrates the experience and judgment of industry authoritative experts and a variety of scientific decision-making algorithm technologies, breaks down cross-departmental and cross-professional engineering management into a quantifiable and traceable indicator system, discovers the hidden correlation between indicators through data mining, and establishes a collaborative relationship between investment plans, milestone plans and funding plans. Compared with the existing full-process management of engineering projects, it has the advantages of high precision, fast speed, and high logical correlation. It can improve the logic of the data layer and use algorithm tools to help project managers discover important control factors, thereby strengthening the collaborative management of the entire process.

[0050] 2. The present invention is highly inclusive and can realize automatic construction functions. For power engineering projects of various sizes and types, combined with objective factors such as specific project management indicators and subjective judgments based on the rich experience of experts in the industry, the present invention can automatically build collaborative relationships and process big data, which can save a lot of manpower and improve data processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the method flow of the present invention.

[0052] Figure 2 It is a schematic diagram of the system module of the present invention. DETAILED DESCRIPTION

[0053] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0054] In order to better illustrate the present embodiment, some parts in the drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product;

[0055] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0056] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0057] Example 1

[0058] This embodiment provides a method for identifying collaborative relationship characteristics of power engineering projects. Figure 1 As shown, the following steps are included:

[0059] S1: Collect information on power projects that have been included in investment plans, implemented, or generated funding payments in the past five years across departments;

[0060] S2: Determine whether the investment plan, milestone plan and funding plan of the collected projects match. If they match, proceed to step S4; if not, proceed to step S3;

[0061] S3: Build a three-plan matching relationship feature index library and output the three-plan matching relationship situation;

[0062] S4: Build a characteristic indicator database of the synergistic relationship between the three plans and output the synergistic relationship between the three plans;

[0063] S5: Integrate the three-plan matching relationship characteristic index library and the three-plan collaborative relationship characteristic index library to output the results.

[0064] Example 2

[0065] This embodiment further discloses the following contents based on the embodiment 1:

[0066] The project information in step S1 includes basic information and process-related data, wherein the basic information includes the project number, project type and implementation unit, and the process-related data includes the node time, investment and payment amount in the project implementation.

[0067] Before step S2, the following steps are also performed:

[0068] Among the projects collected in step S1, unfinished projects and projects with missing information data are eliminated.

[0069] In step S2, it is determined whether the investment plan, milestone plan and funding plan of the collected projects match each other, specifically:

[0070] Using the project number as a unique identifier, summarize the investment plans, milestone plans and funding plans of the past five years, and batch search whether each project exists in the investment plan, milestone plan and funding plan at the same time. If not, the project is a project with management problems, and the three plans do not match, and go to step S3; if so, the three plans of the project match, and go to step S4.

[0071] In step S3, a three-plan matching relationship characteristic index library is constructed, and the three-plan matching relationship is output, specifically:

[0072] S3.1: Calculation of the three-plan matching relationship index database: Combined with the project full-process control stage, through system sorting and historical experience, sort out all relevant factors that affect the three-plan matching and cause management problem projects in the full-process control of engineering projects. For example, projects that have never been included in the investment plan but have generated funding payments are defined as free projects. The reasons for free projects may include: the small project amount leads to the failure of finance to discover it in time, the short implementation time leads to the completion of the project before the project is established, and the system loopholes lead to the failure to verify whether the project is officially established before the funds are paid. Other factors, these relevant factors are quantifiable and calculable indicators, and various indicators are classified according to the management problem projects to form an index database. The indicators are screened through the expert scoring method, and then the indicators of each project that does not match the project database are calculated to form a database;

[0073] S3.2: Indicator matching statistics: Through the association rule set, the implicit and valuable connections between the indicator libraries are mined. The project refers to the various attributes of any non-empty record, and the item set refers to the set of all attributes. All the attributes of all non-empty records are stored in the database D). Assume that there are N project data in the sample space Ω, and each project contains the deviation of M key indicators. The deviation of each indicator in different projects is represented by 0 or 1. The deviation is 1 and the non-deviation is 0. The management rules of the key control indicators of the project can be expressed in three forms: support, confidence and lift. The support refers to the probability that the antecedent X and the consequent Y of the rule in the item set occur at the same time, indicating the probability that the control indicators x and y of the project sample data deviate at the same time. σ(X,Y) represents the total number of times x and y deviate at the same time. N is the number of project samples. The support Support(X→Y) is calculated as follows:

[0074]

[0075]

[0076] Confidence refers to the probability of the occurrence of the rule's subsequent term Y under the premise that the rule's antecedent term X occurs in the term set. It means the probability that the control indicator y will deviate when the control indicator x deviates. The calculation method of confidence (X→Y) is as follows:

[0077]

[0078] Lift refers to the ratio of the probability of the rule-based item Y occurring after the preceding item X in the item set to the probability of the rule-based item Y occurring. It indicates the frequency with which the occurrence of item set X leads to the occurrence of item set Y. The calculation method of lift (X→Y) is as follows:

[0079]

[0080] S3.3: Eliminate matching relationship conclusions that do not meet most of the indicators: Based on the preliminary data calculation results, the thresholds of support, confidence, and lift are set to 0.08, 0.4, and 1, respectively. Association rules that meet the thresholds indicate that the antecedent and consequent of the rules also have a certain positive correlation. Association rules that are lower than the thresholds indicate that the correlation between the antecedent and consequent of the rules is weak, so such association rules are eliminated;

[0081] S3.4: Determination of the characteristic index library of the matching relationship between the three plans: According to the required association rule scale and traceability requirements, adjust the optimization threshold, determine the indicators corresponding to the strongly related association rules, and form the characteristic index library of the matching relationship between the three plans;

[0082] S3.5: Output of the matching relationship between the three plans of the project: Select association rules that meet the support, confidence and lift thresholds, interpret the relationship between the antecedents and the subsequent items of the association rules, and output the matching relationship between the three plans.

[0083] In step S3.2, the direct correlation relationship in the project matching index is analyzed by the Apriori algorithm to mine the frequent item sets of the project data.

[0084] The step S3.2 follows the core idea of ​​multiple rounds of iterations to gradually realize the mining of frequent itemsets, specifically:

[0085] In the first round of iteration, the minimum support threshold is determined and compared, and the candidate set C is selected. 1 Generate frequent item sets L 1 ;

[0086] In subsequent multiple loop iterations, candidate C is generated from the frequent item set index set. k+1 The indicator set is obtained through the frequent item set L k The indicator set is connected to generate candidate C k+1 At the same time, relying on the two properties that any non-empty subset of the frequent indicator set is frequent and any superset of the infrequent indicator set is infrequent, the candidate C is completed by deleting the infrequent item sets in the candidate indicator set. k+1 Pruning of indicator sets;

[0087] The iterative process continues until no new frequent indicator sets can be generated. Then the confidence and lift of all frequent multi-item sets are calculated respectively, and it is determined whether they exceed the threshold. The algorithm outputs all frequent multi-item association rules that meet the requirements.

[0088] In step S3.2, the deviation of each indicator of each project in the mismatched project library is collected by power engineering related management experts through a survey questionnaire method, and statistics are collected. If the management expert believes that a certain indicator deviation causes the project mismatch, it is marked as 1, otherwise it is marked as 0, and the association relationship is mined through the R language program.

[0089] In step S4, a characteristic index library of the three-plan collaborative relationship is constructed to output the three-plan collaborative relationship situation, specifically:

[0090] S4.1: Calculation of the three-plan coordination relationship indicator database. For matching projects that exist in the investment plan, milestone plan, and funding plan at the same time, study whether the corresponding investment completion progress, engineering implementation progress, and funding payment progress during the entire implementation process of the project are coordinated. Therefore, based on the available milestone data, the project full-process control stage is divided into three stages: construction, settlement, and final settlement. Through system sorting, historical experience, and expert scoring methods, sort out all possible relevant factors that affect the coordination of the three plans in the full-process control of engineering projects, form quantifiable and calculable indicators, classify various indicators according to the control stage, form an indicator database, screen the indicators through the expert scoring method, and then calculate the indicators of each project in the project database to form a database;

[0091] S4.2: Indicator matching statistics: Apriori algorithm is used to mine frequent item sets of project data. Through questionnaire survey, the deviation of each indicator of each project in the project library is collected by power engineering management experts to conduct statistics. If the manager believes that a certain indicator deviation causes the project to be mismatched, it is marked as 1, otherwise it is marked as 0. The association relationship is mined through R language program;

[0092] S4.3: Eliminate matching relationship conclusions that do not meet most of the indicators: Based on the preliminary data calculation results, the thresholds of support, confidence, and lift are set to 0.06, 0.3, and 1, respectively. Association rules that meet the thresholds indicate that the antecedent and consequent of the rules also have a certain positive correlation. Association rules that are lower than the thresholds indicate that the correlation between the antecedent and consequent of the rules is weak, so such association rules are eliminated;

[0093] S4.4: Determination of the characteristic index library of the synergistic relationship between the three plans: According to the required association rule scale and traceability requirements, adjust the optimization threshold, determine the indicators corresponding to the strongly related association rules, and form the characteristic index library of the synergistic relationship between the three plans;

[0094] S4.5: Output of the synergistic relationship of the three plans of the project: Select association rules that meet the support, confidence and lift thresholds, interpret the relationship between the antecedent and subsequent items of the association rules, and output the synergistic relationship of the three plans.

[0095] In the specific embodiment, a specific implementation method of the indicator library operation is given, which is specifically:

[0096] Taking the coordination index as an example, through system sorting, the whole process of project management is divided into the feasibility study preliminary design stage, the construction stage, the completion acceptance stage and its sub-stages, and the influencing factors affecting the coordination of the three plans in each stage are sorted out, as shown in Table 1.

[0097] Table 1

[0098]

[0099] For each indicator, its definition should be made clear to avoid deviations when subsequent experts score. For example, the progress of project submission and approval is defined as the time waiting for approval after the project is reported to the higher authorities. If the approval time is long, it will delay the progress of a series of work such as later decision-making, design, and construction. The progress of site selection investigation should take into account that the time spent on site selection investigation may affect the overall decision-making time of the project, thereby affecting the progress of subsequent links; the progress of design document submission is usually within 5 working days after the review. This indicator examines whether the submission node is overdue. Use the expert scoring method to screen it and select the three-plan collaborative key indicator system with representative indicators.

[0100] In addition to the quantifiable indicator data in the engineering management system, sample data of other indicators can be obtained through questionnaire surveys for calculation. Relevant management experts are invited to conduct statistics on the deviations of various indicators based on historical project conditions. The data on indicator deviations are represented in the form of a fact table, with variable values ​​of 1 or 0. 1 means that the manager finds that the indicator has deviated during the construction of a certain project, and 0 means that the indicator has not deviated, forming an indicator activity fact table, as shown in Table 2:

[0101] Table 2

[0102]

[0103] This embodiment also provides a specific implementation method for index matching statistics, which is:

[0104] According to the collected survey forms, after sorting out the effectiveness, the correlation relationship was mined through the R language program, and the thresholds of support, confidence, and lift were set to 0.06, 0.3, and 1 respectively. The calculation results are shown in Table 3:

[0105] Table 3

[0106]

[0107]

[0108] As can be seen from the table, under the conditions of minimum support, confidence, and lift, 12 association rules of the three-plan coordination indicators were mined, and the lift was greater than 1, indicating that the antecedents and the latter of the 12 association rules have a certain positive correlation. Taking the first association rule as an example, the support of "site selection progress → material fulfillment and distribution" is 0.0610, indicating that among all the projects included in the project set, 6.1% of the projects have deviations in the indicators of "site selection progress" and "material fulfillment and distribution". The confidence of 0.7062 means that in the project set, 70.62% of the projects with deviations in "site selection progress" also have deviations in "material fulfillment and distribution". The lift of 1.7249 means that the probability of deviations in "site selection progress" leading to deviations in "material fulfillment and distribution" has increased by 1.731 times.

[0109] This embodiment also provides a specific implementation method for removing weak correlation indicators and outputting correlation relationships and indicator libraries, which is specifically:

[0110] Combining the 12 association rules, it can be seen that the confidence of 5 management rules is greater than 0.5 and the lift is greater than 2, which means that the antecedents and consequents of these 5 association rules have a strong positive correlation, while the antecedents and consequents of the other 7 association rules have a weak correlation.

[0111] According to management needs, association rules with weak correlation are eliminated, and the indicator library corresponding to association rules with strong correlation is sorted out simultaneously.

[0112] Example 3

[0113] This embodiment provides a system for identifying collaborative relationship characteristics of power engineering projects. The system is applied to the method for identifying collaborative relationship characteristics of power engineering projects described in Embodiment 1 and Embodiment 2. Figure 2 As shown, the system includes:

[0114] A collection module, which is used to collect information on projects in power engineering projects that have been included in investment plans, implemented, or generated funding payments in the past five years across departments;

[0115] A matching module, which is used to determine whether the investment plan, milestone plan and funding plan in the collected projects match. If they match, the module enters the matching relationship feature module; if not, the module enters the collaborative relationship feature module;

[0116] A matching relationship feature module, which is used to construct a three-plan matching relationship feature index library and output the three-plan matching relationship status;

[0117] A collaborative relationship feature module, which is used to construct a three-plan collaborative relationship feature index library and output the three-plan collaborative relationship status;

[0118] The output module is used to integrate the three-plan matching relationship feature index library and the three-plan collaborative relationship feature index library to output the results.

[0119] The same or similar reference numerals correspond to the same or similar components;

[0120] The terms used in the drawings to describe positional relationships are only used for illustrative purposes and should not be construed as limiting this patent;

[0121] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for identifying collaborative relationship characteristics of power engineering projects, characterized in that: The following steps are involved: S1: Collect information on power projects that have been included in investment plans, implemented, or generated funding payments in the past five years across departments; S2: Determine whether the investment plan, milestone plan and funding plan of the collected projects match. If they match, proceed to step S4; if not, proceed to step S3; S3: Build a three-plan matching relationship feature index library and output the three-plan matching relationship situation; S4: Build a characteristic indicator database of the synergistic relationship between the three plans and output the synergistic relationship between the three plans; S5: integrate the three-plan matching relationship characteristic index library and the three-plan collaborative relationship characteristic index library, and output the results; In step S2, it is determined whether the investment plan, milestone plan and funding plan of the collected projects match each other, specifically: Using the project number as a unique identifier, summarize the investment plans, milestone plans and funding plans of the past five years, and search in batches whether each project exists in the investment plan, milestone plan and funding plan at the same time. If not, the project is a project with management problems, and the three plans do not match, and go to step S3; If yes, the three plans of the project are matched, and the process goes to step S4; In step S3, a three-plan matching relationship characteristic index library is constructed, and the three-plan matching relationship is output, specifically: S3.1: Calculation of the three-plan matching relationship index library: Combined with the project full process control stage, through system sorting and historical experience, sort out all relevant factors that affect the three-plan matching and cause management problem projects in the full process control of engineering projects, and classify these relevant factors into quantifiable and calculable indicators according to the management problem projects to form an index library. The indicators are screened through the expert scoring method, and then the indicators of each project in the unmatched project library are calculated to form a database; S3.2: Indicator matching statistics: Through the association rule set, the implicit and valuable connections between the indicator libraries are mined. The project refers to the various attributes of any non-empty record, and the item set refers to the set of all attributes. All the attributes of all non-empty records are stored in the database D). Assume that there are N project data in the sample space Ω, and each project contains the deviation of M key indicators. The deviation of each indicator in different projects is represented by 0 or 1. The deviation is 1 and the non-deviation is 0. The management rules of the key control indicators of the project can be expressed in three forms: support, confidence and lift. The support refers to the probability that the antecedent X and the consequent Y of the rule in the item set occur at the same time, indicating the probability that the control indicators x and y of the project sample data deviate at the same time. σ(X,Y) represents the total number of times x and y deviate at the same time. N is the number of project samples. The support Support(X→Y) is calculated as follows: Confidence refers to the probability of the occurrence of the rule term Y under the premise that the rule term X occurs in the term set. It means the probability of the control indicator y also deviating when the control indicator x deviates. The calculation method of confidence (X→Y) is as follows: Lift refers to the ratio of the probability of the rule-based item Y occurring after the preceding item X in the item set to the probability of the rule-based item Y occurring. It indicates the frequency with which the occurrence of item set X leads to the occurrence of item set Y. The calculation method of lift (X→Y) is as follows: S3.3: Eliminate matching relationship conclusions that do not meet most of the indicators: Based on the preliminary data calculation results, the thresholds of support, confidence, and lift are set to 0.08, 0.4, and 1, respectively. Association rules that meet the thresholds indicate that the antecedent and consequent of the rules also have a certain positive correlation. Association rules that are lower than the thresholds indicate that the correlation between the antecedent and consequent of the rules is weak, so such association rules are eliminated; S3.4: Determination of the characteristic index library of the matching relationship between the three plans: According to the required association rule scale and traceability requirements, adjust the optimization threshold, determine the indicators corresponding to the strongly related association rules, and form the characteristic index library of the matching relationship between the three plans; S3.5: Output of the matching relationship between the three plans of the project: Select the association rules that meet the support, confidence and lift thresholds, interpret the relationship between the before and after items of the association rules, and output the matching relationship between the three plans; In step S4, a characteristic index library of the three-plan collaborative relationship is constructed to output the three-plan collaborative relationship situation, specifically: S4.1: Calculation of the three-plan coordination relationship indicator database. For matching projects that exist in the investment plan, milestone plan, and funding plan at the same time, study whether the corresponding investment completion progress, engineering implementation progress, and funding payment progress during the entire implementation process of the project are coordinated. Therefore, based on the available milestone data, the project full-process control stage is divided into three stages: construction, settlement, and final settlement. Through system sorting, historical experience, and expert scoring methods, sort out all possible relevant factors that affect the coordination of the three plans in the full-process control of engineering projects, form quantifiable and calculable indicators, classify various indicators according to the control stage, form an indicator database, screen the indicators through the expert scoring method, and then calculate the indicators of each project in the project database to form a database; S4.2: Indicator matching statistics: Apriori algorithm is used to mine frequent item sets of project data. Through questionnaire survey, the deviation of each indicator of each project in the project library is collected by power engineering management experts to conduct statistics. If the manager believes that a certain indicator deviation causes the project to be mismatched, it is marked as 1, otherwise it is marked as 0. The association relationship is mined through R language program; S4.3: Eliminate matching relationship conclusions that do not meet most of the indicators: Based on the preliminary data calculation results, the thresholds of support, confidence, and lift are set to 0.06, 0.3, and 1, respectively. Association rules that meet the thresholds indicate that the antecedent and consequent of the rules also have a certain positive correlation. Association rules that are lower than the thresholds indicate that the correlation between the antecedent and consequent of the rules is weak, so such association rules are eliminated; S4.4: Determination of the characteristic index library of the synergistic relationship between the three plans: According to the required association rule scale and traceability requirements, adjust the optimization threshold, determine the indicators corresponding to the strongly related association rules, and form the characteristic index library of the synergistic relationship between the three plans; S4.5: Output of the synergistic relationship of the three plans of the project: Select association rules that meet the support, confidence and lift thresholds, interpret the relationship between the antecedent and subsequent items of the association rules, and output the synergistic relationship of the three plans.

2. The method for identifying collaborative relationship characteristics of power engineering projects according to claim 1, characterized in that: The project information in step S1 includes basic information and process-related data, wherein the basic information includes the project number, project type and implementation unit, and the process-related data includes the node time, investment and payment amount in the project implementation.

3. The method for identifying collaborative relationship characteristics of power engineering projects according to claim 2, characterized in that: The step S2 is preceded by the following steps: Among the projects collected in step S1, unfinished projects and projects with missing information data are eliminated.

4. The method for identifying collaborative relationship characteristics of power engineering projects according to claim 3 is characterized in that: In step S3.2, the direct correlation relationship in the project matching index is analyzed by the Apriori algorithm to mine the frequent item sets of the project data.

5. The method for identifying collaborative relationship characteristics of power engineering projects according to claim 4, characterized in that: The step S3.2 follows the core idea of ​​multiple rounds of iterations to gradually realize the mining of frequent itemsets, specifically: In the first round of iteration, the minimum support threshold is determined and compared to generate the frequent item set L1 from the candidate item set C1; In subsequent multiple loop iterations, candidate C is generated from the frequent item set index set. k+1 The indicator set is obtained through the frequent item set L k The indicator set is connected to generate candidate C k+1 At the same time, relying on the two properties that any non-empty subset of the frequent indicator set is frequent and any superset of the infrequent indicator set is infrequent, the candidate C is completed by deleting the infrequent item sets in the candidate indicator set. k+1 Pruning of indicator sets; The iterative process continues until no new frequent indicator sets can be generated. Then the confidence and lift of all frequent multi-item sets are calculated respectively, and it is determined whether they exceed the threshold. The algorithm outputs all frequent multi-item association rules that meet the requirements.

6. The method for identifying collaborative relationship characteristics of power engineering projects according to claim 5, characterized in that: In step S3.2, the deviation of each indicator of each project in the mismatched project library is collected by power engineering related management experts through a survey questionnaire method, and statistics are collected. If the management expert believes that a certain indicator deviation causes the project mismatch, it is marked as 1, otherwise it is marked as 0, and the association relationship is mined through the R language program.

7. A power engineering project collaborative relationship feature identification system, characterized in that: The feature identification system is applied to the method for identifying features of collaborative relationships of power engineering projects according to any one of claims 1 to 6, and the system comprises: A collection module, which is used to collect information on projects in power engineering projects that have been included in investment plans, implemented, or generated funding payments in the past five years across departments; A matching module, which is used to determine whether the investment plan, milestone plan and funding plan in the collected projects match. If they match, the module enters the matching relationship feature module; if not, the module enters the collaborative relationship feature module; A matching relationship feature module, which is used to construct a three-plan matching relationship feature index library and output the three-plan matching relationship status; A collaborative relationship feature module, which is used to construct a three-plan collaborative relationship feature index library and output the three-plan collaborative relationship status; The output module is used to integrate the three-plan matching relationship feature index library and the three-plan collaborative relationship feature index library to output the results.

Images