A full-process management system for electric power smart engineering projects
Through the full-process management system of the power smart engineering project, the problem that traditional project management methods cannot meet the needs of modern power smart engineering projects is solved, comprehensive control of the project and risk prediction are achieved, and management efficiency and task completion quality are improved.
Patent Information
- Application Number
- CN202411233224.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-04
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-04
AI Technical Summary
Traditional project management methods cannot meet the needs of modern power smart engineering projects with high precision, strong real-time and multi-party collaborative cooperation, resulting in low project management efficiency and difficulty in risk prediction.
A full-process management system for power smart engineering projects was designed, adopting highly integrated modular design, including project establishment, progress management, cost management, contract management, procurement management, and report and data display modules, using databases and blockchains With intelligent algorithms, we conduct real-time data updates and analysis, intelligent recommendation task allocation and progress prediction.
It realizes comprehensive control and management of projects, ensures real-time updates and accuracy of data, improves project management efficiency, effectively predicts and avoids risks, and ensures that tasks are assigned to the most appropriate members and projects are successfully completed.
Smart Images

Figure CN118735473B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of project monitoring, and in particular to a full-process management system for electric power smart engineering projects. Background Art
[0002] In the management of modern power smart projects, the complexity and scale of projects are increasing year by year. With the continuous improvement of the demand for project management, traditional project management methods can no longer meet the current needs of high precision, strong real-time performance and multi-party collaboration. Therefore, building a full-process management system based on intelligence, dataization, and automation is crucial to the success of power smart project projects. Summary of the invention
[0003] In order to solve the above problems, the purpose of the present invention is to provide a full-process management system for electric power smart engineering projects. Through a highly integrated modular design, comprehensive control and management of the project is achieved, and real-time updating, transparency and accuracy of data are guaranteed throughout the entire process from project initiation to final completion.
[0004] To achieve the above object, the present invention adopts the following technical solutions:
[0005] A full-process management system for electric power smart engineering projects, characterized by comprising a project establishment module, a database, a progress management module, a cost management module, a contract management module, a procurement management module and a report and data display module;
[0006] Project establishment module, used for project approval, project basic information entry and WBS initialization, and storing the generated project and task information in the database;
[0007] The progress management module generates a detailed task list for the project based on the project and task information in the database, and intelligently recommends task allocation plans based on the historical performance and current workload of project members; based on the progress prediction model, it predicts the project progress and completion time and generates a prediction curve;
[0008] The cost management module formulates detailed budgets according to tasks and project stages, tracks real-time project costs, compares them with budgets, generates cost curves, and analyzes real-time data during project execution. Based on time series analysis models, it predicts future cost situations and warns of overspending risks in advance.
[0009] The contract management module stores contracts based on blockchain, tracks the progress of contract execution, and automatically updates the contract status in conjunction with the progress management module;
[0010] The procurement management module automatically generates purchase orders, tracks procurement progress, and is integrated with the progress management module and cost management module;
[0011] The report and data display module summarizes the data of each module, builds dynamic reports and dashboards, and provides support for the overall progress and cost control of the project.
[0012] Further, initialize the WBS as follows:
[0013] Confirm the highest level of the WBS, usually the project name or main deliverable, and break down the tasks level by level;
[0014] Further decompose each major deliverable into smaller work packages until you reach a granular task that cannot be decomposed any further;
[0015] And design a structured data structure to store WBS, including tasks at all levels of the project and their attributes;
[0016] Use auto-increment ID to generate a unique ID for each WBS node and the duration of each parent task From the duration of the subtask Adding up:
[0017] ;
[0018] Among them, m is the number of subtasks, n is the parent task, For the subtasks;
[0019] The start and end dates are calculated based on the project's overall calendar and each task's dependencies:
[0020] ;
[0021] ;
[0022] in, is the start time of the parent task, For the The start time of each subtask; The end time of the parent task. For the The end time of each subtask;
[0023] The level of each task starts at 0, the root task is level 0, its subtasks are level 1, and so on.
[0024] Furthermore, the data structure should include the following fields: WBS_ID: uniquely identifies each WBS node; Project_ID: unique identifier associated with the project; Task_Name: task name; Parent_ID: ID of the parent task; Level: WBS level; Duration: time required for the task; Start_Date: task start date; End_Date: task end date; Resources: resource requirements.
[0025] Furthermore, based on the project and task information in the database, a detailed task list of the project is generated, and a task allocation plan is intelligently recommended based on the historical performance and current workload of project members, as follows:
[0026] Extract task information and project member information from the database, including historical performance and current workload;
[0027] Calculate each member's adjusted performance according to the adjusted performance formula;
[0028] Adjusted_Performance(u)=History_Performance(u)×(1-Current_Workload(u))
[0029] Where: Adjusted_Performance(u) is the adjusted performance of project member u; History_Performance(u) is the historical performance score of project member u; Current_Workload(u) is the current workload of project member u;
[0030] According to the task requirements and the skills of the members, calculate the fitness matrix Suitability(i,u);
[0031] Suitability(i,u)=Adjusted_Performance(u)×Task_Fit(i,u);
[0032] Among them, Suitability(i,u) represents the suitability of member u for task i; Task_Fit(i,u) is the matching degree between task i and member u;
[0033] Use the Hungarian algorithm to find the most suitable task allocation solution, where:
[0034] The overall optimization goal is:
[0035] ;
[0036] in, is a binary variable, indicating whether task i is assigned to member u;
[0037] Constraints:
[0038] ;
[0039] ;
[0040] Among them, Task_Duration(i) is the duration of task i, Max_Workload(u) is the maximum workload of member u;
[0041] The optimized task allocation results are stored in the task allocation table.
[0042] Furthermore, the Hungarian algorithm is used to find the task allocation solution with the highest suitability, as follows:
[0043] ① Preprocess the fitness matrix Suitability(i,u) to ensure that the algorithm can work properly, including
[0044] Row subtraction operation, subtracting the minimum value of each row from each element of each row, so that each row has at least one zero;
[0045] Column subtraction operation: If there are still columns without zeros, subtract the minimum value of each column from each element of each column so that each column has at least one zero;
[0046] ② Use the least number of horizontal and vertical lines to cover all zero elements in the matrix. If the number of lines used is equal to the dimension of the matrix, the optimal allocation has been found; otherwise, go to step ③.
[0047] ③ Adjust the uncovered elements of the matrix:
[0048] Find the minimum value among the uncovered elements;
[0049] Subtract this minimum value from all uncovered elements;
[0050] Add the minimum value to all elements that are covered twice;
[0051] Then return to step ② and re-cover the zero elements;
[0052] Finally, the tasks and members corresponding to the zero elements in the fitness matrix are the optimal allocation solutions.
[0053] Furthermore, the progress prediction model is built based on CatBoost, as follows:
[0054] Obtain historical data from the database, including data related to completed projects and tasks, and perform feature extraction to obtain task features and time features;
[0055] Generate feature interaction terms to capture the relationship between features:
[0056] X interaction =X1×X2;
[0057] Among them, X1 and X2 are the original features, X interaction Indicates the interactive characteristics of the two;
[0058] Encoding categorical features using CatBoost’s built-in methods (e.g. Task_Name),
[0059] X encoded =CatBoostEncoder(X categorical );
[0060] Among them, CatBoostEncoder represents the built-in encoding of CatBoost; X categorical Represents the category feature; X encoded Represents the encoded category features;
[0061] And use the CatBoost model to predict task duration:
[0062] Predicted_Duration=F(X)
[0063] Among them, the input feature moment X includes numerical features X numeric , interactive feature X interaction and the encoded categorical feature X encoded, F() represents the CatBoost model.
[0064] Furthermore, the CatBoost regression model is as follows:
[0065] Initialize the model by predicting the mean of the target variable:
[0066]
[0067] Where L is the loss function, It is The true value of the sample, F0(X) is the initial predicted value; N is the number of samples; c is the constant value determined by minimizing the loss function in the initial state;
[0068] Iteratively construct the gradient tree, in At iteration:
[0069] Calculate residuals :
[0070] ;
[0071] in, Predicted value for the current prediction model;
[0072] Training a new tree To fit the residuals:
[0073] ;
[0074] in, Represents the tree h for the sample The predicted value of
[0075] Update the model predictions:
[0076] ;
[0077] in, is the learning rate.
[0078] Furthermore, a detailed budget is prepared according to the tasks and project stages, and the real-time cost of the project is tracked and compared with the budget to generate a cost curve, as follows:
[0079] Before the project starts, prepare a detailed budget based on project requirements and stages, broken down by tasks, subtasks and project stages;
[0080] Collect and record this data by entering the actual cost of each task as it is performed.
[0081] Compare tracked actual costs to budget to identify variances and generate visual cost curves.
[0082] Furthermore, based on the blockchain storage contract, the progress of contract execution is tracked, and the progress management module is combined to automatically update the contract status, as follows:
[0083] Store the contract file in IPFS and obtain the hash value;
[0084] Create a smart contract instance and record the contract hash and initial state to the blockchain;
[0085] Monitor smart contract events through blockchain nodes, including contract creation and status updates; parse event data and synchronize it to the database;
[0086] When the contract status changes, the progress management module is notified through the API interface. The progress management module automatically updates the project progress according to the contract status to ensure data consistency.
[0087] Furthermore, the report and data display module includes a data collection layer, a data processing layer, a data storage layer and a data display layer, wherein:
[0088] Data collection layer, collects data from the contract management module, project progress management module, and cost control module;
[0089] The data processing layer cleans, aggregates, and analyzes the collected data to generate intermediate results;
[0090] Data storage layer, which stores the processed data in the database;
[0091] Data presentation layer, using data visualization tools to present data, generate dynamic reports and dashboards,
[0092] The dynamic report displays dynamic reports including the number of contracts, contract status distribution, project task completion status, cost budget and actual expenditure data. The report supports filtering and querying by time, project, contract status and cost type dimensions;
[0093] The dashboard displays key indicators of the project through charts and provides an intuitive view of the overall situation of the project.
[0094] The present invention has the following beneficial effects:
[0095] 1. The present invention realizes comprehensive control and management of projects through highly integrated modular design. From project establishment to final completion, it ensures real-time updating, transparency and accuracy of data. Through intelligent algorithms and data analysis models, the system can not only improve the efficiency of project management, but also effectively predict and avoid risks.
[0096] 2. The present invention uses the Hungarian algorithm based on the fitness matrix to find the task allocation scheme with the highest fitness, which not only takes into account the historical performance and current workload of project members, but also achieves the optimal task matching, ensuring that tasks are assigned to the most suitable members, thereby maximizing the execution efficiency of the project and the quality of task completion;
[0097] 3. The present invention effectively predicts the project progress and completion time based on CatBoost and generates prediction curves. Through these predictions, the project progress can be better understood, potential construction period risks can be identified, and corresponding countermeasures can be taken to ensure the smooth completion of the project. BRIEF DESCRIPTION OF THE DRAWINGS
[0098] Figure 1 This is a system architecture diagram of the present invention. DETAILED DESCRIPTION
[0099] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:
[0100] refer to Figure 1 In this embodiment, a full-process management system for electric power smart engineering projects is provided, including a project establishment module, a database, a progress management module, a cost management module, a contract management module, a procurement management module, and a report and data display module;
[0101] Project establishment module, used for project approval, project basic information entry and WBS initialization, and storing the generated project and task information in the database;
[0102] The progress management module generates a detailed task list for the project based on the project and task information in the database, and intelligently recommends task allocation plans based on the historical performance and current workload of project members; based on the progress prediction model, it predicts the project progress and completion time and generates a prediction curve;
[0103] The cost management module formulates detailed budgets according to tasks and project stages, tracks real-time project costs, compares them with budgets, generates cost curves, and analyzes real-time data during project execution. Based on time series analysis models, it predicts future cost situations and warns of overspending risks in advance.
[0104] The contract management module stores contracts based on blockchain, tracks the progress of contract execution, and automatically updates the contract status in conjunction with the progress management module;
[0105] The procurement management module automatically generates purchase orders, tracks procurement progress, and is integrated with the progress management module and cost management module;
[0106] The report and data display module summarizes the data of each module, builds dynamic reports and dashboards, and provides support for the overall progress and cost control of the project.
[0107] In this embodiment, the WBS is initialized as follows:
[0108] Confirm the highest level of the WBS, usually the project name or main deliverables (such as design, implementation, testing, etc.), and decompose the tasks level by level;
[0109] Further decompose each major deliverable into smaller work packages until you reach a granular task that cannot be decomposed any further;
[0110] And design a structured data structure to store WBS, including tasks at all levels of the project and their attributes;
[0111] Use auto-increment ID to generate a unique ID for each WBS node and the duration of each parent task From the duration of the subtask Adding up:
[0112] ;
[0113] Among them, m is the number of subtasks, n is the parent task, For the subtasks;
[0114] The start and end dates are calculated based on the project's overall calendar and each task's dependencies:
[0115] ;
[0116] ;
[0117] in, is the start time of the parent task, For the The start time of each subtask; The end time of the parent task. For the The end time of each subtask;
[0118] The level of each task starts at 0, the root task is level 0, its subtask is level 1, and so on.
[0119] In this embodiment, the data structure should include the following fields: WBS_ID: unique identifier of each WBS node; Project_ID: unique identifier associated with the project; Task_Name: task name; Parent_ID: ID of the parent task; Level: WBS level; Duration: time required for the task; Start_Date: task start date; End_Date: task end date; Resources: resource requirements.
[0120] In this embodiment, based on the project and task information in the database, a detailed task list of the project is generated, and based on the historical performance and current workload of the project members, a task allocation plan is intelligently recommended, as follows:
[0121] Extract task information and project member information from the database, including historical performance and current workload;
[0122] Calculate each member's adjusted performance according to the adjusted performance formula;
[0123] Adjusted_Performance(u)=History_Performance(u)×(1-Current_Workload(u))
[0124] Where: Adjusted_Performance(u) is the adjusted performance of project member u; History_Performance(u) is the historical performance score of project member u; Current_Workload(u) is the current workload of project member u;
[0125] According to the task requirements and the skills of the members, calculate the fitness matrix Suitability(i,u);
[0126] Suitability(i,u)=Adjusted_Performance(u)×Task_Fit(i,u);
[0127] Among them, Suitability(i,u) represents the suitability of member u for task i; Task_Fit(i,u) is the matching degree between task i and member u (considering the member's skills and task requirements);
[0128] Use the Hungarian algorithm to find the most suitable task allocation solution, where:
[0129] The overall optimization goal is:
[0130] ;
[0131] in, is a binary variable, indicating whether task i is assigned to member u;
[0132] Constraints:
[0133] ;
[0134] ;
[0135] Among them, Task_Duration(i) is the duration of task i, Max_Workload(u) is the maximum workload of member u;
[0136] The optimized task allocation results are stored in the task allocation table.
[0137] In this embodiment, the Hungarian algorithm is used to find the task allocation solution with the highest suitability, as follows:
[0138] ① Preprocess the fitness matrix Suitability(i,u) to ensure that the algorithm can work properly, including
[0139] Row subtraction operation, subtracting the minimum value of each row from each element of each row, so that each row has at least one zero;
[0140] Column subtraction operation: If there are still columns without zeros, subtract the minimum value of each column from each element of each column so that each column has at least one zero;
[0141] ② Use the least number of horizontal and vertical lines to cover all zero elements in the matrix. If the number of lines used is equal to the dimension of the matrix, the optimal allocation has been found; otherwise, go to step ③.
[0142] ③ Adjust the uncovered elements of the matrix:
[0143] Find the minimum value among the uncovered elements;
[0144] Subtract this minimum value from all uncovered elements;
[0145] Add the minimum value to all elements that are covered twice;
[0146] Then return to step ② and re-cover the zero elements;
[0147] Finally, the tasks and members corresponding to the zero elements in the fitness matrix are the optimal allocation solutions.
[0148] In this embodiment, the progress prediction model is built based on CatBoost, as follows:
[0149] Obtain historical data from the database, including data related to completed projects and tasks, and perform feature extraction to obtain task features and time features;
[0150] Among them, task characteristics include task type, planned duration, number of predecessor tasks, task complexity, and task importance; time characteristics include planned start time (converted to timestamp), planned end time (converted to timestamp), seasonal factors (month, quarter, etc.); resource characteristics include resource type (Resource_Type),
[0151] Resource utilization (Resource_Utilization), experience and skill level of project members.
[0152] Generate feature interaction terms to capture the relationship between features:
[0153] X interaction =X1×X2;
[0154] Among them, X1 and X2 are the original features, X interaction Indicates the interactive characteristics of the two;
[0155] For example, X1 represents "estimated hours" (Estimated_Hours), and X2 represents "resource cost" (Resource_Cost), then the interaction feature Xinteraction is interpreted as "the product of hours and resource cost". This feature can provide additional information to help the model understand the impact of hours on resource cost.
[0156] Encoding categorical features using CatBoost’s built-in methods (e.g. Task_Name),
[0157] X encoded =CatBoostEncoder(X categorical );
[0158] Among them, CatBoostEncoder represents the built-in encoding of CatBoost; X categorical Represents the category feature; X encoded Represents the encoded category features;
[0159] And use the CatBoost model to predict task duration:
[0160] Predicted_Duration=F(X)
[0161] Among them, the input feature moment X includes numerical features X numeric , interactive feature X interaction and the encoded categorical feature X encoded, F() represents the CatBoost model.
[0162] In this embodiment, the CatBoost regression model is as follows:
[0163] Initialize the model by predicting the mean of the target variable:
[0164] ;
[0165] Where L is the loss function, It is The true value of the sample, F0(X) is the initial predicted value; N is the number of samples; c is the constant value determined by minimizing the loss function in the initial state;
[0166] Iteratively construct the gradient tree, in At iteration:
[0167] Calculate residuals :
[0168] ;
[0169] in, Predicted value for the current prediction model;
[0170] Training a new tree To fit the residuals:
[0171] ;
[0172] in, Represents the tree h for the sample The predicted value of
[0173] Update the model predictions:
[0174] ;
[0175] in, is the learning rate.
[0176] In this embodiment, a detailed budget is prepared according to the tasks and project stages, the real-time cost of the project is tracked, and the cost is compared with the budget to generate a cost curve, as follows:
[0177] Before the project starts, prepare a detailed budget based on project requirements and stages, broken down by tasks, subtasks and project stages;
[0178] Collect and record this data by entering the actual cost of each task as it is performed.
[0179] Compare tracked actual costs to budget to identify variances and generate visual cost curves.
[0180] In this embodiment, the contract is stored based on blockchain, and the progress of contract execution is tracked. The contract status is automatically updated in conjunction with the progress management module, as follows:
[0181] Store the contract file in IPFS and obtain the hash value;
[0182] Create a smart contract instance and record the contract hash and initial state to the blockchain;
[0183] Monitor smart contract events through blockchain nodes, including contract creation and status updates; parse event data and synchronize it to the database;
[0184] When the contract status changes, the progress management module is notified through the API interface. The progress management module automatically updates the project progress according to the contract status to ensure data consistency.
[0185] In this embodiment, the report and data display module includes a data collection layer, a data processing layer, a data storage layer and a data display layer, wherein:
[0186] Data collection layer, collects data from the contract management module, project progress management module, and cost control module;
[0187] The data processing layer cleans, aggregates, and analyzes the collected data to generate intermediate results;
[0188] Data storage layer, which stores the processed data in the database;
[0189] Data presentation layer, using data visualization tools to present data, generate dynamic reports and dashboards,
[0190] The dynamic report displays dynamic reports including the number of contracts, contract status distribution, project task completion status, cost budget and actual expenditure data. The report supports filtering and querying by time, project, contract status and cost type dimensions;
[0191] The dashboard displays key indicators of the project (such as progress, cost, risk, etc.) through charts, providing an intuitive view of the overall project situation.
[0192] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0193] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0194] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1A function specified in one or more boxes.
[0195] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0196] The above is only a preferred embodiment of the present invention, and does not limit the present invention in other forms. Any technician familiar with the profession may use the above disclosed technical content to change or modify it into an equivalent embodiment with equivalent changes. However, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention still belongs to the protection scope of the technical solution of the present invention.
Claims
1. A full-process management system for electric power smart engineering projects, characterized in that: It includes project establishment module, database, progress management module, cost management module, contract management module, procurement management module and report and data display module; Project establishment module, used for project approval, project basic information entry and WBS initialization, and storing the generated project and task information in the database; The progress management module generates a detailed task list for the project based on the project and task information in the database, and intelligently recommends task allocation plans based on the historical performance and current workload of project members; based on the progress prediction model, it predicts the project progress and completion time and generates a prediction curve; The cost management module formulates detailed budgets according to tasks and project stages, tracks real-time project costs, compares them with budgets, generates cost curves, and analyzes real-time data during project execution. Based on time series analysis models, it predicts future cost situations and warns of overspending risks in advance. The contract management module stores contracts based on blockchain, tracks the progress of contract execution, and automatically updates the contract status in conjunction with the progress management module; The procurement management module automatically generates purchase orders, tracks procurement progress, and is integrated with the progress management module and cost management module; Report and data display module, which aggregates data from each module, builds dynamic reports and dashboards, and provides support for overall project progress and cost control; Initialize WBS as follows: Confirm the highest level of the WBS, including the main deliverables, and decompose the tasks level by level; Further decompose each major deliverable into work packages until you reach a granular level where you can no longer decompose it; And design a structured data structure to store WBS, including the various levels of tasks and their attributes of the project; Use auto-increment ID to generate a unique ID for each WBS node and the duration of each parent task From the duration of the subtask Adding up: ; Among them, m is the number of subtasks, n is the parent task, For the subtasks; The start and end dates are calculated based on the project's overall calendar and each task's dependencies: ; ; in, is the start time of the parent task, For the The start time of each subtask; The end time of the parent task. For the The end time of each subtask; The level of each task starts from 0, the root task is level 0, its subtask is level 1, and so on; The data structure should include the following fields: WBS_ID: uniquely identifies each WBS node; Project_ID: unique identifier associated with the project; Task_Name: task name; Parent_ID: ID of the parent task; Level: WBS level; Duration: time required for the task; Start_Date: task start date; End_Date: task end date; Resources: resource requirements; The progress prediction model is built based on CatBoost, as follows: Obtain historical data from the database, including data related to completed projects and tasks, and perform feature extraction to obtain task features and time features; Generate feature interaction terms to capture the relationship between features: X interaction =X1×X2; Among them, X1 and X2 are the original features, X interaction Indicates the interactive characteristics of the two; Use CatBoost's built-in method to encode categorical features. X encoded =CatBoostEncoder(X categorical ); Among them, CatBoostEncoder represents the built-in encoding function of CatBoost; X categorical Represents the category feature; X encoded Represents the encoded category features; And use the CatBoost model to predict task duration: Predicted_Duration=F(X) Among them, the input feature matrix X includes numerical features X numeric , interactive feature X interaction and the encoded categorical feature X encoded, F() represents the CatBoost model; The CatBoost model is as follows: Initialize the model by predicting the mean of the target variable: ; Where L is the loss function, It is The true value of the sample, F0(X) is the initial predicted value; N is the number of samples; c is the constant value determined by minimizing the loss function in the initial state; Iteratively construct the gradient tree, in At iteration: Calculate residuals : ; in, Predicted value for the current prediction model; Training a new tree To fit the residuals: ; in, Represents the tree h for the sample The predicted value of Update the model predictions: ; in, is the learning rate.
2. The full-process management system of a power smart engineering project according to claim 1 is characterized in that: Based on the project and task information in the database, a detailed task list of the project is generated, and a task allocation plan is intelligently recommended based on the historical performance and current workload of the project members, as follows: Extract task information and project member information from the database, including historical performance and current workload; Calculate each member's adjusted performance according to the adjusted performance formula; Adjusted_Performance(u)=History_Performance(u)×(1-Current_Workload(u)) Where: Adjusted_Performance(u) is the adjusted performance of project member u; History_Performance(u) is the historical performance score of project member u; Current_Workload(u) is the current workload of project member u; According to the task requirements and the skills of the members, calculate the fitness matrix Suitability(i,u); Suitability(i,u)=Adjusted_Performance(u)×Task_Fit(i,u); Among them, Suitability(i,u) represents the suitability of member u for task i; Task_Fit(i,u) is the matching degree between task i and member u; Use the Hungarian algorithm to find the most suitable task allocation solution, where: The overall optimization goal is: ; in, is a binary variable, indicating whether task i is assigned to member u; Constraints: ; ; Among them, Task_Duration(i) is the duration of task i, Max_Workload(u) is the maximum workload of member u; The optimized task allocation results are stored in the task allocation table.
3. The full-process management system of a power smart engineering project according to claim 2 is characterized in that: The Hungarian algorithm is used to find the task allocation solution with the highest fitness, as follows: ① Preprocess the fitness matrix Suitability(i,u) to ensure that the algorithm can work properly, including Row subtraction operation, subtracting the minimum value of each row from each element of each row, so that each row has at least one zero; Column subtraction operation: If there are still columns without zeros, subtract the minimum value of each column from each element of each column so that each column has at least one zero; ② Use the least number of horizontal and vertical lines to cover all zero elements in the matrix. If the number of lines used is equal to the dimension of the matrix, the optimal allocation has been found; otherwise, go to step ③; ③ Adjust the uncovered elements of the matrix: Find the minimum value among the uncovered elements; Subtract this minimum value from all uncovered elements; Add the minimum value to all elements that are covered twice; Then return to step ② and re-cover the zero elements; Finally, the tasks and members corresponding to the zero elements in the fitness matrix are the optimal allocation solutions.
4. The full-process management system of a power smart engineering project according to claim 1 is characterized in that: The detailed budget is formulated according to the tasks and project stages, the real-time cost of the project is tracked, and the cost curve is generated by comparing it with the budget, as follows: Before the project starts, prepare a detailed budget based on project requirements and stages, broken down by tasks, subtasks and project stages; Collect and record this data by entering the actual cost of each task as it is performed. Compare tracked actual costs to budget to identify variances and generate visual cost curves.
5. The full-process management system of a power smart engineering project according to claim 1 is characterized in that: The blockchain-based contract storage and tracking of contract execution progress are combined with the progress management module to automatically update the contract status, as follows: Store the contract file in IPFS and obtain the hash value; Create a smart contract instance and record the contract hash and initial state to the blockchain; Monitor smart contract events through blockchain nodes, including contract creation and status updates; parse event data and synchronize it to the database; When the contract status changes, the progress management module is notified through the API interface. The progress management module automatically updates the project progress according to the contract status to ensure data consistency.
6. The full-process management system of a power smart engineering project according to claim 1 is characterized in that: The report and data display module includes a data collection layer, a data processing layer, a data storage layer and a data display layer, wherein: Data collection layer, collects data from the contract management module, project progress management module, and cost control module; The data processing layer cleans, aggregates, and analyzes the collected data to generate intermediate results; Data storage layer, which stores the processed data in the database; Data presentation layer, using data visualization tools to present data, generate dynamic reports and dashboards, The dynamic report displays dynamic reports including the number of contracts, contract status distribution, project task completion status, cost budget and actual expenditure data. The report supports filtering and querying by time, project, contract status and cost type dimensions; The dashboard displays key indicators of the project through charts and provides an intuitive view of the overall situation of the project.
Citation Information
Cited By
Project risk prediction avoidance method and system fused with deep learning
CN121032197A
Project risk prediction avoidance method and system fusing deep learning
CN121032197B
