Water conservancy project construction management informatization platform

Abstract

The invention discloses a water conservancy project construction management informatization platform, and relates to the technical field of water conservancy project management, and the platform comprises an evaluation form generation module, a form filling verification module, an intelligent approval module, a classification filing module and a safety management module. According to the invention, a whole-user, full-service and full-life-cycle construction management standardization system is constructed; automatic filling and real-time verification are realized by means of a natural language processing technology and machine learning, and data entry experience and accuracy are improved; decision support is provided through big data analysis, a filling strategy and an approval path are recommended through a model, and process management is optimized; digitized upgrading of form full-life-cycle management is realized, and the method is suitable for a water conservancy project construction management scene with a complex approval process; the state evolution trend of hidden danger points or major hazard sources in a future period of time is simulated, and early warning and plan generation are carried out based on the state evolution trend.

Description

Technical Field

[0001] The present invention relates to the technical field of hydropower and water conservancy project management, and in particular to an information platform for water conservancy project construction management. Background Art

[0002] The construction management information platform is a system that provides digital, networked, and intelligent management for the entire life cycle of construction projects, providing comprehensive support and services for engineering construction projects. Because water conservancy projects often involve complex environments, large investments, and long construction cycles, project management requirements are often more stringent. The application of an information platform in water conservancy projects can significantly improve project management and execution efficiency.

[0003] Traditional project management platforms, such as Chinese patent application No. 201910870408.7, disclose an integrated EPC project management platform, which is logically divided into a display layer, an application layer, and a data layer from a technical implementation perspective. The EOS platform is used as a development tool, and a database is used to store and manage data. The software can run on the middleware of the operating system, and can aggregate design data and project management data to achieve auxiliary planning and design capabilities.

[0004] However, traditional EPC project management platforms still have the following shortcomings: 1. The business scope is not comprehensive, lacking business functions such as design management and supervision management, and cannot meet the needs of digital management of the entire project life cycle. 2. There is a lack of process management tools, and the system still has two sets of online and offline systems.

[0005] Currently, no effective solutions have been proposed for the problems in related technologies. Summary of the Invention

[0006] In response to the problems in the related technologies, the present invention proposes an information platform for water conservancy project construction management to overcome the above-mentioned technical problems existing in the existing related technologies.

[0007] To this end, the specific technical solutions adopted in the present invention are as follows: A water conservancy project construction management information platform, comprising: Evaluation form generation module, used to generate and adjust the quality evaluation form of water conservancy projects based on predefined templates and business rules; The form filling and verification module is used to automatically fill in the quality inspection and evaluation form; based on the verification rules of regular expressions and logistic regression models, the validity and consistency of the content of the quality inspection and evaluation form are verified; The intelligent approval module is used to identify the target clauses of the quality inspection and review form using the target clause recognition model; construct a graph based on the target clauses and convert the graph into structured approval decision factors; and generate approval recommendations based on the approval decision factors and historical approval data. The classification and archiving module is used to implement classification and archiving management of quality inspection and evaluation forms using text classification algorithms; The safety management module is used to predict the state evolution trend of water conservancy project risk factors using chaos testing methods and fuzzy cognitive maps, and to establish early warning response mechanisms and response plans.

[0008] Furthermore, based on predefined templates and business rules, the quality inspection and evaluation forms for water conservancy projects are generated and adjusted, including: Based on document processing library technology, combined with predefined templates and business rules, it automatically generates quality inspection and evaluation forms for water conservancy projects; Based on the rule engine and using contextual information including user role and project stage, the content of the quality review form is dynamically adjusted.

[0009] Furthermore, the automatically filled quality inspection form includes: Obtain historical data on water conservancy project quality inspections and evaluations, and select natural language processing technology and machine learning algorithms; Use natural language processing technology and machine learning algorithms to identify data patterns and user habits in historical data, and fill in quality inspection forms based on the obtained data patterns and user habits.

[0010] Furthermore, based on the verification rules of regular expressions and logistic regression models, the validity and consistency of the content of the quality review form are verified, including: Perform rule validation on the user's input data for quality assessment, and return error information corresponding to the input data that does not conform to the regular expression rules to the user; After the input data passes the regular expression rule verification, the logistic regression verification model is used to analyze the validity and consistency of the input data, and when the input data meets the data validity and consistency, a success message is returned to the user.

[0011] Furthermore, using the target clause identification model, the target clauses of the quality inspection form are identified as follows: The language model is trained using the form data with the target terms labeled, and is jointly trained with the conditional random field layer to optimize the language model parameters and obtain the target term recognition model. The text extraction technology is used to extract the form data to be analyzed, and the form data to be analyzed is substituted into the target clause recognition model to obtain the target clause of the quality inspection form.

[0012] Furthermore, a graph is constructed based on the target terms and converted into structured approval decision factors, including: Determine the semantic elements of the target clauses based on the type of quality inspection and evaluation form and the regulatory requirements of the water conservancy project; Determine the relationship between the semantic elements of the target clauses, use the semantic elements as nodes and the relationships between semantic elements as edges, and use graph construction technology to build a graph; Using data structuring technology, the graph is converted into structured approval decision factors.

[0013] Furthermore, using text classification algorithms, the quality inspection and evaluation forms are classified and archived, including: Using historical form data, we trained the support vector machine text classification algorithm module; Preprocess the data of the quality inspection form, and substitute the preprocessed form data into the support vector machine text classification algorithm module, and combine the preset classification model and rules to output the category of the quality inspection form; Through the archiving execution module, the quality inspection and evaluation forms are archived according to their categories.

[0014] Furthermore, using the chaos test method and fuzzy cognitive map, the state evolution trend of water conservancy project risk factors is predicted, including: Obtain construction site time series data for water conservancy projects; perform chaos testing and phase space reconstruction on the construction site time series data, and build a local prediction model by optimizing prediction parameters to achieve short-term prediction of the time series of various risk factors of water conservancy projects; A fuzzy cognitive map of each risk factor node is constructed, and combined with the short-term prediction results of the time series of each risk factor of water conservancy project, the node status of each risk factor is updated, and the risk status matrix of water conservancy project construction in the future is generated.

[0015] Furthermore, the establishment of early warning response mechanisms and response plans includes: Based on the risk status matrix, combined with water conservancy project industry standards and historical experience, early warning thresholds are set, and risk levels of water conservancy project hazards are determined, and multi-level early warning information is output; Through plan matching rules and based on different risk levels and risk scenarios, corresponding dynamic plans are generated.

[0016] Furthermore, a water conservancy project construction management information platform also includes an intelligent analysis module for visualizing and personalizing water conservancy project construction data based on a big data platform.

[0017] The beneficial effects of the present invention are: (1) A standardized construction management system applicable to all users, all businesses, and the entire life cycle with the project legal person as the core has been constructed. Different from the traditional single-perspective management model, this invention combines the current practice of water conservancy project construction management, takes the work task dimension as the main content line, focuses on the construction management life cycle dimension from all aspects and all elements, and takes the construction user dimension as the coverage scope to construct a 1+3+N construction management standard system, that is, a construction management standard system architecture with the project legal person as the core, including the three core management dimensions of "life cycle", "work task" and "construction user" and the content of "N" construction management tasks, and can continuously supplement, update and improve the content of the standardized system according to actual needs.

[0018] (2) Based on the standardization system, a construction management information platform was constructed to meet the needs of the project legal person and the on-site management organization at two levels of construction management, and to achieve the management goals of all bidding sections, all participating units, all management businesses, and the entire life cycle. Unlike the ineffective information construction that ignores the role of standardization and is superficial, the present invention embeds the constructed standardization system and results into the design of the construction management information platform. The platform design follows the concept of "comprehensive content, advanced technology, reasonable design, and moderate advancement", which can meet the actual needs of the project legal person and the on-site management organization. It includes seven levels and all management business modules. By introducing advanced information technology and adopting the "one-time development, gradual upgrade" model, the efficient construction and complete functions of the information platform are achieved.

[0019] (3) The electronic form intelligent parsing and dynamic generation technology in this invention solves the problems of manual development, repetitive labor, and multi-source data processing by automatically generating forms, thereby improving the efficiency of form creation. It also uses natural language processing technology and machine learning to achieve automatic filling and real-time verification, improving the data entry experience and accuracy. At the same time, it uses big data analysis to provide decision support, relies on models to recommend filling strategies and approval paths, and optimizes process management. In terms of process automation and intelligent approval, it automatically triggers the approval process and provides intelligent approval suggestions to reduce manual intervention. Finally, through intelligent classification and archiving and advanced search functions, it improves the efficiency of form management and retrieval, and realizes the digital upgrade of form management throughout its life cycle. It is suitable for water conservancy project construction management scenarios with complex approval processes.

[0020] (4) Based on fuzzy cognitive maps and chaotic time series prediction, this invention constructs a dynamic risk evolution model to simulate the state evolution trend of hidden danger points or major hazard sources over a period of time in the future, and based on this, it conducts early warning and generates emergency plans. Compared with traditional static assessment methods, this method has stronger foresight, adaptability and dynamic response capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0022] Figure 1 This is a flow chart of a project quality inspection and evaluation form according to an embodiment of the present invention; Figure 2 is a form verification flow chart according to an embodiment of the present invention; Figure 3 is a form filling flow chart according to an embodiment of the present invention; Figure 4 This is a form approval process according to an embodiment of the present invention; Figure 5 This is a user query process according to an embodiment of the present invention; Figure 6 is a flow chart of water conservancy project safety management according to an embodiment of the present invention; Figure 7 This is the architecture diagram of the water conservancy project construction management information platform; Figure 8 The figure is a principle block diagram of a water conservancy project construction management information platform according to an embodiment of the present invention.

[0023] In the picture: 1. Assessment form generation module; 2. Form filling and verification module; 3. Intelligent approval module; 4. Classification and archiving module; 5. Security management module. DETAILED DESCRIPTION

[0024] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. By referring to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0025] According to an embodiment of the present invention, a water conservancy project construction management information platform is provided.

[0026] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 8 As shown, the water conservancy project construction management information platform according to an embodiment of the present invention includes: Assessment form generation module 1 is used to generate and adjust the quality inspection and evaluation form of water conservancy projects based on predefined templates and business rules; form filling and verification module 2 is used to automatically fill in the quality inspection and evaluation form; verification rules based on regular expressions and logistic regression models are used to verify the validity and consistency of the content of the quality inspection and evaluation form; intelligent approval module 3 is used to use the target clause identification model to identify the target clauses of the quality inspection and evaluation form; a map is constructed according to the target clauses, and the map is converted into a structured approval decision factor; approval suggestions are generated according to the approval decision factors and historical approval data, and the approval suggestions are returned to the user interaction layer for display to the user; classification and archiving module 4 is used to implement classification and archiving management of quality inspection and evaluation forms using text classification algorithms; safety management module 5 is used to use chaos test method and fuzzy cognitive map to predict the state evolution trend of water conservancy project risk factors, and establish early warning response mechanism and response plan.

[0027] In one embodiment, generating and adjusting a water conservancy project quality inspection and evaluation form based on a predefined template and business rules includes: Based on document processing library technology, combined with predefined templates and business rules, the quality inspection and evaluation form for water conservancy projects is automatically generated; based on the rule engine and utilizing contextual information including user roles and project stages, the content of the quality inspection and evaluation form is dynamically adjusted.

[0028] In one embodiment, automatically filling in the quality review form includes: Obtain historical data on water conservancy project quality inspection and evaluation, and select natural language processing technology and machine learning algorithms; use natural language processing technology and machine learning algorithms to identify data patterns and user habits in historical data, and fill in the quality inspection and evaluation form based on the obtained data patterns and user habits.

[0029] In one embodiment, based on the verification rules of regular expressions and logistic regression models, the validity and consistency verification of the content of the quality review form includes: The user's input data for quality assessment is subject to rule verification, and the error information corresponding to the input data that does not comply with the regular expression rules is returned to the user; after the input data passes the regular expression rule verification, the input data is analyzed for validity and consistency using a logistic regression verification model, and when the input data meets the data validity and consistency, a success message is returned to the user.

[0030] In one embodiment, the target clause identification model is used to identify the target clauses of the quality inspection form, including: The language model is trained using form data with marked target clauses, and is jointly trained with the conditional random field layer to optimize the parameters of the language model and obtain a target clause recognition model. The form data to be analyzed is extracted using text extraction technology, and the form data to be analyzed is substituted into the target clause recognition model to obtain the target clauses of the quality inspection form.

[0031] In one embodiment, constructing a graph based on target terms and converting the graph into structured approval decision factors includes: According to the type of quality inspection and evaluation form and the specification requirements of water conservancy projects, the semantic elements of the target clauses are determined; the relationship between the semantic elements of the target clauses is determined, and a graph is constructed using graph construction technology, with the semantic elements as nodes and the relationships between semantic elements as edges; and data structuring technology is used to convert the graph into structured approval decision factors.

[0032] In one embodiment, using a text classification algorithm to implement classified archiving management of quality inspection and evaluation forms includes: Using historical form data, a support vector machine text classification algorithm module is trained; the data of the quality inspection form is preprocessed, and the preprocessed form data is substituted into the support vector machine text classification algorithm module, and the category of the quality inspection form is output in combination with the preset classification model and rules; through the archiving execution module, the quality inspection form is archived according to the category of the quality inspection form.

[0033] In one embodiment, using the chaos test method and fuzzy cognitive map, the state evolution trend of water conservancy project risk factors is predicted, including: Acquire construction site time series data of water conservancy projects; perform chaos test and phase space reconstruction on the construction site time series data, and build a local prediction model by optimizing prediction parameters to achieve short-term prediction of the time series of various risk factors of water conservancy projects; construct a fuzzy cognitive map of each risk factor node, and combine the short-term prediction results of the time series of various risk factors of water conservancy projects to update the node status of each risk factor and generate a risk status matrix for water conservancy project construction in the future.

[0034] In one embodiment, establishing an early warning response mechanism and a response plan includes: Based on the risk status matrix, combined with water conservancy project industry standards and historical experience, warning thresholds are set, and risk levels of water conservancy project hazards are judged, and multi-level warning information is output; through plan matching rules, and according to different risk levels and risk scenarios, corresponding dynamic plans are generated.

[0035] In one embodiment, a water conservancy project construction management information platform also includes an intelligent analysis module for visualizing and performing personalized modeling on water conservancy project construction data based on a big data platform.

[0036] In order to facilitate understanding of the above technical solutions of the present invention, the working principle of the present invention in actual process is described in detail below.

[0037] Water conservancy project construction involves many disciplines and is difficult to manage. In addition to the dynamic management of the core business of project construction management from the five traditional aspects of technical support, safety prevention and control, investment savings, progress control, and reliable quality, its water conservancy project construction management information platform also needs to meet the functional requirements of diversified business, mobile office, data aggregation, decision support, consultation support, and command issuance, so as to achieve the goal of improving construction management level and management efficiency and strengthening the coordination of management business.

[0038] The business requirements of the water conservancy project construction management information platform are divided into the following categories according to different business stages: Preliminary management: Construction units can carry out preliminary management work in the system in accordance with relevant regulations on water conservancy project construction management and in compliance with laws and regulations, including the preparation, submission, review, consultation and approval of project proposals and feasibility study reports, as well as related special tasks such as environmental impact assessment and preliminary work on soil and water conservation, to lay the foundation for project construction.

[0039] Design Management: Design management mainly involves comprehensive management of the design unit's drawing supply plan, design drawings, design results, design changes, etc.

[0040] Drawing Delivery Plan Management: Based on the overall project schedule, design milestones, and the needs of the decomposed tasks, we develop work plans for the design progress and drawing delivery. This includes the development, execution, and tracking of design schedules. It also synchronizes the design plan with the overall schedule, enabling automatic updates and early warnings to ensure smooth project progress.

[0041] Design drawing management: Register and manage design results and design drawing information. After the design unit produces the drawing, the system supports initiating a construction drawing review process. According to the construction review process, the construction drawing is submitted to the construction unit for online review. The review process, review opinions, responses, and implementation status are recorded. Problems discovered during the review process are immediately fed back to the design unit. The design unit can make modifications to the original drawing as needed, upload the updated version, and re-enter the review process until the construction drawing review is passed.

[0042] Design results management: Results release management is mainly used to register and manage construction drawing release information. The system supports distributing approved construction drawings to construction teams, supervision units, etc. through online processes.

[0043] Design change management: register and manage basic information, design change reports, review opinions, etc. for general and major design changes caused by reasons such as actual conditions not being consistent with the original design during construction, technical improvements, material changes, changes in construction conditions, etc.

[0044] Bidding and procurement: It realizes the supervision of bidding and procurement systems and classified cataloging for consulting services, engineering construction, material and equipment procurement and other related types. At the same time, it supports project legal persons to draft, initiate, review, countersign and approve the preliminary results of bidding planning work such as the type classification of various bidding contents, section division and contract package determination online, and compare the progress of bidding work with its corresponding plan.

[0045] (4) Construction commencement management: Online filling and review of construction commencement registration forms and their accompanying attachments and materials, and obtaining process guidance and reminders for key construction commencement management tasks.

[0046] (5) Contract management: Realize online contract planning management, contract document management, contract process management and contract ledger management from preliminary design to final acceptance.

[0047] Contract planning management: According to relevant regulations, the construction unit must decompose the design unit project into several sub-projects and corresponding contracts based on the preliminary design of the project and its budget approval, the structural characteristics of the project, the current state of the construction market, and the requirements of relevant policies and regulations. These include contracts for engineering design, engineering construction, equipment procurement, consulting services, labor leasing, and other contracts. The system also determines the bidding and signing time for each contract project based on the overall progress plan of the design unit project. The actual progress of the contract planning is compared online with the planned progress in a variety of vivid and intuitive ways, such as curves, bar charts, and bar graphs. On-time, delayed, or ahead of schedule are presented in different colors, lines, and icons. A list of major contract planning tasks and a corresponding list of work results are established. Completed and unfinished contract planning work results are represented in different colors to demonstrate the progress of the work at that stage. Reminders and warnings are issued for major contract planning tasks and key nodes.

[0048] Contract document management: supports the drafting, preparation, countersigning and approval of various contract documents, and the final result is an electronic document; supports the simultaneous display of the signed contract text and the electronic document, and mutual comparison.

[0049] Contract process management: coding, storage, review, retrieval and related work management of contract documents.

[0050] Contract ledger management: record the process of issuing, borrowing and calling the contract text, form ledger information, and store it in the system database.

[0051] (6) Investment Management: Investment target management: The construction unit determines the responsible department and, based on the total investment amount determined in the approved feasibility study report and under the overall goal of the project, decomposes and compiles the investment targets for each stage, individual projects and unit projects and other cost control targets within the system. In accordance with management regulations and procedures, these are discussed, reviewed and submitted for approval. These will serve as the cost targets for subsequent work and project implementation, guide the orderly and compliant implementation of various tasks, and ensure the realization of investment management goals.

[0052] Budget management: Based on the approved preliminary design budget, in accordance with the principles of total control, reasonable adjustment, and matching with the progress, according to the actual construction management and investment control management authority, the execution budget is prepared in accordance with the prescribed procedures, and the execution budget is adjusted regularly according to the actual progress of the project.

[0053] Investment Plan Management: Based on construction needs and project schedules, monthly, quarterly, and annual investment plans are compiled, submitted, issued, and completed based on budget estimates. Funds are supervised and dispatched within the system, including application, receipt, payment, and confirmation. Construction units should integrate and coordinate the special funding plans compiled by participating units to form a comprehensive construction funding plan.

[0054] (7) Quality management: Implement online quality management tasks such as quality system establishment, project division, quality inspection and assessment, quality inspection, quality testing, quality supervision, and quality defect filing.

[0055] Supervision and construction units can conduct online quality inspection and assessment through mobile terminals and web terminals, and support online filling, review and approval of quality assessment forms. After passing, electronic signature documents will be automatically generated, and quality inspection and assessment materials can be archived with one click. For online quality inspection and assessment, first add an inspection and assessment form template, use the preset Excel sample form as the inspection and assessment form, select the preset template, configure the quantity and mount it to the specified project division node, fill in the quality inspection and assessment form for approval and electronic signature, and complete the quality inspection and assessment operation. Figure 1 As shown, the quality inspector fills out the inspection and evaluation form, the supervisor reviews the inspection and evaluation form, and the system administrator configures the inspection and evaluation form.

[0056] Leveraging Apache POI (an open-source Java document processing library), the system automatically generates customized forms based on predefined templates and business rules. For multi-source, multi-attribute data, the system can quickly convert .XLSX and .XLSX data files into dynamic HTML (Hypertext Markup Language) forms through a user-friendly interface. In addition to enabling real-time form content adjustments, the system further develops a rules engine that dynamically adjusts form content based on contextual information such as user role and project phase. This process not only accelerates form design and deployment but also efficiently transforms data from electronic forms into interactive information and data flows, providing users with a personalized and dynamic interactive experience. Combining natural language processing and machine learning algorithms, the system can identify patterns and user habits in historical data and automatically populate form content, reducing manual entry and improving data accuracy and efficiency. Furthermore, the system integrates advanced validation mechanisms, including validation rules based on regular expressions and logistic regression models, to verify the validity and consistency of user input in real time. This ensures data entry accuracy and compliance instantly, compared to traditional methods.

[0057] For example: (a) Collect structured form data, semi-structured logs, and unstructured text, and convert them to a standardized format. User profile features are extracted, field relationships are analyzed, and missing and outlier values ​​are corrected using business-defined rules and statistical methods. (b) Semantic understanding and field classification based on natural language processing are used to analyze field meanings and determine their types using pre-trained language models. Semantic dependencies between fields are captured using sequence models for contextual analysis. User behavior modeling and personalized habit capture based on machine learning are used to learn commonly used values ​​and preferred ordering from historical user records. (c) Auto-fill: Directly reuse user-repeated historical values; predict the most likely value using a Bayesian model; infer implicit information through knowledge graph associations; and automatically select the optimal solution based on user historical preferences when there are conflicting values. Based on machine learning models and analysis results and historical data, the system can recommend optimal form filling strategies or approval paths to users, helping them identify best practices and optimization points for common processes, thereby improving overall project management efficiency. The core machine learning models include: (1) Sequence Recommendation Model: Transformer / LSTM analyzes the filling order pattern and predicts the efficient field filling path; (2) Graph Neural Network (GNN): Constructs the "form field-approval node" association graph and learns the optimal approval link. After the form is submitted, the system automatically triggers the corresponding approval process according to the approval rules and conditions preset in the water conservancy project supervision specifications. Based on the form content and historical data, it predicts the most likely approval node sequence and recommends a "risk avoidance path": avoiding the approval node combination with a high rejection rate in history. It also supports complex approval chains and conditional branches by integrating tools such as rule engine parsing, greatly reducing manual intervention and improving the efficiency and transparency of approval.

[0058] like Figure 2 As shown in the figure, after a user enters data in the user interaction layer, the data (form data) first enters the regular expression validation module. If the input data conforms to the regular expression rules, it enters the logistic regression validation module for further verification. If it does not conform to the regular expression rules, the validation logic layer immediately returns an error message (indicating that the input format is incorrect) to the user interaction layer, alerting the user to the incorrect input. The logistic regression validation module analyzes the data based on a trained model. If the data is deemed valid and consistent, a success message (indicating that the input is valid) is returned to the user interaction layer. If the data is deemed to contain a problem, an error message (indicating that the input format is incorrect) is also returned to the user interaction layer.

[0059] like Figure 3 As shown, the data enters the data preprocessing module through the data input layer and undergoes feature engineering. After using the pattern recognition model, semantic association graph, and automatic filling engine, the form is filled and the user behavior log is recorded, or error prompts and corrections are given to collect user feedback.

[0060] After the form is submitted, the system will automatically trigger the corresponding approval process based on the approval rules and conditions preset in the water conservancy project supervision specifications, greatly reducing manual intervention and improving the efficiency and transparency of approval.

[0061] The system then integrates the BERT+CRF model to identify the core terms in the form, builds a semantic element graph of the form, converts unstructured text into structured approval decision factors, and provides approval recommendations based on historical approval data, implementing an intelligent approval mechanism. Compared to traditional methods, this can speed up the approval process while maintaining consistency and accuracy in decision-making. The specific method is as follows: After receiving the form data, the data access layer passes it to the text extraction module in the core processing layer. The extracted text enters the BERT+CRF model recognition module, which identifies the core terms and outputs the results. A large amount of form data with labeled core terms is collected and divided into training, validation, and test sets. Using BERT (a pre-trained language model) as the base model, fine-tuning is performed on the training set and combined with a CRF layer (conditional random field layer) for joint training to optimize model parameters. The trained model is evaluated using the test set and then deployed in the recognition module in the core processing layer.

[0062] The semantic element graph construction module constructs a graph based on core clauses, and the data structuring module then converts the graph into structured approval decision factors. Semantic elements are determined based on the form type and water conservancy project specifications. Based on the core clauses identified by the BERT+CRF model, relevant semantic elements are extracted and the relationships between elements are analyzed. A graph database is used to construct a form semantic element graph, with semantic elements as nodes and relationships between elements as edges, to intuitively present form semantic information.

[0063] The intelligent decision-making layer obtains decision factors and historical approval data, generates approval suggestions after analysis, and returns the suggestions to the user interaction layer for display to the user.

[0064] like Figure 4 As shown, the form is submitted and the form data is parsed by the rule engine. Based on the conditions, the multi-level approval process or single-level approval process is entered, approval tasks are assigned by role, and the approver is notified.

[0065] After a form is submitted, it is automatically classified and filed based on its content and metadata. By applying the SVM (Support Vector Machine) text classification algorithm, the system can automatically file the form into the corresponding category or folder for easy subsequent management and retrieval. The specific method is as follows: After the user submits form data in the user interaction layer, the data is passed to the form preprocessing module in the classification processing layer. This module performs preprocessing operations such as cleaning the data and extracting key information, and then sends the processed data to the SVM text classification algorithm module.

[0066] The SVM text classification algorithm module collects a large amount of historical form data, annotates it according to different categories, and constructs training and test datasets. It performs word segmentation and word vector conversion on the text data, extracts text features, and uses the annotated training dataset to train the SVM text classification algorithm. The trained model is then deployed in the SVM text classification algorithm module.

[0067] The SVM text classification algorithm module analyzes the form data based on the preset classification model and rules, predicts its category, and passes the classification results to the archiving execution module. Based on the classification results, the archiving execution module stores the form in the corresponding category or folder in the data storage layer and feeds back the archiving success information to the user interaction layer for display.

[0068] like Figure 5 As shown in the figure, when a user makes a query, the query type is judged, and full-text index matching, structured query conditions, vector space similarity calculation, etc. are performed to obtain a preliminary result set, and the results are sorted and integrated.

[0069] The electronic form intelligent parsing and dynamic generation technology in the present invention solves the problems of manual development repetitive labor and multi-source data processing by automatically generating forms, thereby improving the efficiency of form creation; it uses natural language processing technology and machine learning to achieve automatic filling and real-time verification, improving the data entry experience and accuracy. At the same time, big data analysis is used to provide decision support, and the model is used to recommend filling strategies and approval paths to optimize process management. In terms of process automation and intelligent approval, the approval process is automatically triggered, and intelligent approval suggestions are provided to reduce manual intervention. Finally, through intelligent classification and archiving and advanced retrieval functions, the efficiency of form management and retrieval is improved, and the digital upgrade of the full life cycle management of forms is realized, which is suitable for water conservancy project construction management scenarios with complex approval processes.

[0070] (8) Safety Management: Realize real-time understanding of safety management goals, organizations, systems, training, plans, costs, assessments, etc., realize hidden danger detection and major hazard sources, build a dynamic risk evolution model based on fuzzy cognitive maps and chaotic time series prediction, simulate the state evolution trend of hidden danger points or major hazard sources in the future, and generate early warnings and plans based on this. Compared with traditional static assessment methods, it has stronger foresight, adaptability and dynamic response capabilities. The specific methods are as follows: like Figure 6As shown in the figure, through the sensor network, geological exploration equipment, construction management system, etc. at the water conservancy project construction site, multi-source data such as geological parameters, hydrological and meteorological data, and construction data are collected in real time to provide basic information for the model.

[0071] The collected time series data, such as water level, flow, and construction progress, are first tested for chaos. Phase space reconstruction is then performed using the delayed embedding method. A local prediction model is constructed by optimizing prediction parameters to achieve short-term predictions for the time series of various risk factors. The fuzzy cognitive map (including causal network initialization) is synchronously input into the state iteration calculation phase along with the chaotic prediction results. The iterative formula of the fuzzy cognitive map updates the states of each risk factor node, generating a risk state matrix for the water conservancy project construction over the next period of time. Based on the risk state matrix and warning thresholds set based on water conservancy project industry standards and historical experience, risk levels are determined and multi-level warning information is output. Furthermore, through matching plan rules, dynamic emergency plans are generated for different risk levels and scenarios, including flood control material deployment and personnel evacuation plans for flood risks, and reinforcement construction plans for slope instability risks.

[0072] Traditional static warning thresholds and fixed emergency plans are unable to adapt to environmental changes and dynamic adjustments to construction schedules during water conservancy project construction. The proposed model dynamically updates warning thresholds based on the iterative results of the fuzzy cognitive map, automatically lowering flood risk warning thresholds during flood season. Furthermore, by simulating the risk suppression effects of different intervention measures, it generates more targeted dynamic emergency plans, improving risk response efficiency.

[0073] (9) Progress Management: This system manages and controls the progress of specific individual projects and project groups during the construction implementation phase. Through visual progress management, the system compares and analyzes the completion status with the plan, and promptly corrects any deviations. It also supports online reporting of major project quantities and the corresponding contract amounts by each participating unit.

[0074] (10) Archives management: Realize online collection, compilation and pre-filing of project files, connect with the project legal person's archives management system through the interface, and transfer them to the archives management system with one click.

[0075] (11) Comprehensive management: realize online management of news, notices, project events, policies, regulations, rules and regulations, address books, sending and receiving documents, attendance, meetings, tasks, and awards for excellence, etc.

[0076] (12) Acceptance Management: Realize online acceptance plan management, unit project acceptance management, contract project completion acceptance management, unit trial operation acceptance management, underwater acceptance management, special acceptance management, and completion acceptance management, including acceptance application, acceptance meeting and acceptance appraisal functions.

[0077] (13) Intelligent Analysis: The intelligent analysis module is a cloud-native data analysis service based on an innovative middle-tier architecture. Based on the capabilities of the big data platform, it provides data analysis and visualization for water conservancy project construction, supports convenient integration and application with business systems, and provides self-service analysis and reporting capabilities. It supports users' personalized modeling in a visual design environment and realizes ad hoc chart analysis and exploration based on browsing. With a new scenario-based and immersive experience, the intelligent analysis module meets the data intelligent analysis needs of business systems and provides graphical data visualization management and operation services. From big data development, big data integration and analysis, and data warehouse to data mart construction, it adapts to various scenarios and environments through hot-swappable different cores. The intelligent analysis module supports personalized analysis functions such as reports, charts, analysis, dashboards, and report generation, and has the ability to quickly build digital large screens. It provides multi-data source management, including mainstream relational databases and domain business metadata, and supports the construction of intelligent analysis of data models, chart cards, dashboards, free reports, data mining, prediction and warning, message push, report subscription, tenant permission management, etc.

[0078] (14) Mobile applications: Enable mobile and remote office work, solve complex problems such as delayed message reception and difficulty in collecting engineering management data, and thus realize mobile project management.

[0079] like Figure 7 The following is an architecture diagram of a water conservancy project construction management information platform, which specifically includes: IoT Perception Layer: By deploying IoT devices such as hosts, smart terminals, on-site acoustic sensors, air environment sensors, identification equipment, surveillance cameras, dust suppression equipment, excavators, and cranes, we comprehensively collect structured, unstructured, and real-time data generated during the construction process. These devices provide multi-dimensional perception across the construction site, ensuring data diversity and timeliness, and providing reliable basic data support for project management.

[0080] Data layer: Users can manually enter data through mobile devices or PCs, and sensors, cameras, and other devices can automatically collect and upload data. Data is then stored and categorized according to pre-set storage formats using database technologies such as SQL Server and Oracle, as well as file storage technologies. Through analysis and integrated processing, various database types are formed, including enterprise, project, business, and model libraries. The enterprise library contains user information, organizational information, other business-related static enumeration information, and dynamic information such as corporate systems and processes. The project library contains basic project information, information about participating units, and project documentation. The business library includes project business information such as design management, safety management, and quality management. The model library includes pump station BIM models (Building Information Models), geographic information data, and manually modeled models. Data includes engineering data, enterprise data, and project data.

[0081] Network layer: By establishing dedicated video lines, business networks, mobile communications, dedicated internet lines, and VPNs, combined with 5G technology, we ensure efficient and reliable data transmission. This multi-layered network architecture provides a stable communications foundation, supporting the real-time transmission of multi-source, heterogeneous data for project construction management, effectively preventing information delays or loss.

[0082] The supporting layer includes electronic signature, visualization model, workflow engine, message engine, report engine, and unified identity authentication.

[0083] Business application layer: Based on a standardized information platform, it adopts a three-dimensional matrix functional design, integrates the standard specification system with information technology innovation, and creates a "data-knowledge-management" strongly coupled, multi-level, multi-user full-cycle management model. It designs cross-disciplinary and multi-objective business processes for the management needs of different participating units, clarifies core management standards such as quality, safety, progress, investment, and contracts, and proposes key decision-making nodes, decision-making knowledge, supporting data, and related relationships in business processes. It completes the coupling and connection of business management processes, decision-making knowledge, and data resources, and realizes strong collaboration in water conservancy project construction management. Among them, project pre-stage and preparation management includes pre-stage management and commencement management; project implementation management includes design management, contract management, quality management, commencement management, bidding and procurement, investment management, safety management, and progress management; project acceptance and evaluation management includes acceptance management; and project comprehensive management includes mobile applications, archive management, party building management, and comprehensive services.

[0084] Decision-making and Control Layer: All construction parties collaborate via PC or mobile devices. For example, if a user initiates a quality inspection task on PC, other authorized users can provide feedback via mobile devices, breaking down time and space constraints and improving communication efficiency. Furthermore, the system supports diverse application scenarios such as large-screen data display and QR code scanning, meeting the multi-level and multi-angle management needs of construction parties and enabling comprehensive presentation and convenient operation of project information. The decision-making and control layer includes progress analysis, quality analysis, investment analysis, and risk analysis.

[0085] To meet the practical needs of large-scale water conservancy project construction management, a three-dimensional matrix-style construction management standardization system centered on the project legal person has been established. Led by the standardization system and focused on overcoming key technologies, a multi-subject, full-process, and multi-task construction management information platform has been constructed. This platform enables rapid development, intelligent management, and visual collaboration, resolving the technical and management challenges currently faced by water conservancy projects, which involve a wide range of disciplines and numerous participating organizations. The achievements of this project's construction management standardization and informatization have been successfully applied in projects such as the recent construction of the flood detention area surrounding Hongze Lake in Sihong County, a key provincial water conservancy project, the second phase of the Huaihe River estuary waterway project, the Jiangsu section of the East Route of the South-to-North Water Diversion Project, the old Yellow River course in Suqian City, and the Bosten Lake Pumping Station in Xinjiang. Through the application of the "standardization + informatization" system of construction management, the standardization of engineering construction management has been greatly improved, the coordination ability of the construction management team has been enhanced, and all aspects of project construction management have been made more standardized, efficient and controllable, which has greatly avoided the waste of resources and cost increases caused by information asymmetry, insufficient experience of construction management personnel, and incomplete professional knowledge. In addition, the use of paperless and informatized office forms has reduced the management costs of on-site personnel travel, vehicles, and other office expenses such as printing, copying and mailing of materials. From December 2022 to the present, the results have saved a total of 10 million yuan in costs in the application of various water conservancy projects. The results of this project solved the common bottleneck problems faced by water conservancy project construction management, and have a high industry promotion and application value. It can promote and drive related basic research and industrial development, and promote technological progress in the industry. The promotion and application of research results in similar water conservancy project construction projects will achieve greater economic and social benefits.

[0086] In this context, EPC (Engineering Procurement and Construction) is a project delivery model in which the contractor is responsible for the design, procurement, construction, and commissioning of the project. BIM (Building Information Modeling) is a three-dimensional model that integrates information from the entire lifecycle of a construction project using digital technology to support collaborative work and decision-making.

[0087] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A water conservancy project construction management information platform, characterized by: include: Evaluation form generation module, used to generate and adjust the quality evaluation form of water conservancy projects based on predefined templates and business rules; The form filling and verification module is used to automatically fill in the quality inspection and evaluation form; based on the verification rules of regular expressions and logistic regression models, the validity and consistency of the content of the quality inspection and evaluation form are verified; Intelligent approval module, used to identify the target clauses of the quality inspection form using the target clause recognition model; Construct a graph based on target terms and convert it into structured approval decision factors; generate approval recommendations based on approval decision factors and historical approval data; The classification and archiving module is used to implement classification and archiving management of quality inspection and evaluation forms using text classification algorithms; The safety management module is used to predict the state evolution trend of water conservancy project risk factors using chaos testing methods and fuzzy cognitive maps, and to establish early warning response mechanisms and response plans.

2. A water conservancy project construction management information platform according to claim 1, characterized in that: The generation and adjustment of the water conservancy project quality inspection and evaluation form based on the predefined template and business rules includes: Based on document processing library technology, combined with predefined templates and business rules, it automatically generates quality inspection and evaluation forms for water conservancy projects; Based on the rule engine and using contextual information including user role and project stage, the content of the quality review form is dynamically adjusted.

3. A water conservancy project construction management information platform according to claim 1, characterized in that: The automatic filling quality evaluation form includes: Obtain historical data on water conservancy project quality inspections and evaluations, and select natural language processing technology and machine learning algorithms; Use natural language processing technology and machine learning algorithms to identify data patterns and user habits in historical data, and fill in quality inspection forms based on the obtained data patterns and user habits.

4. A water conservancy project construction management information platform according to claim 1, characterized in that: The verification rules based on regular expressions and logistic regression models for verifying the validity and consistency of the content of the quality inspection form include: Perform rule validation on the user's input data for quality assessment, and return error information corresponding to the input data that does not conform to the regular expression rules to the user; After the input data passes the regular expression rule verification, the logistic regression verification model is used to analyze the validity and consistency of the input data, and when the input data meets the data validity and consistency, a success message is returned to the user.

5. The water conservancy project construction management information platform according to claim 1, characterized in that: The target clause identification model is used to identify the target clauses of the quality inspection form, including: The language model is trained using the form data with the target terms labeled, and is jointly trained with the conditional random field layer to optimize the language model parameters and obtain the target term recognition model. The text extraction technology is used to extract the form data to be analyzed, and the form data to be analyzed is substituted into the target clause recognition model to obtain the target clause of the quality inspection form.

6. The water conservancy project construction management information platform according to claim 1, characterized in that: The process of constructing a graph based on target terms and converting the graph into structured approval decision factors includes: Determine the semantic elements of the target clauses based on the type of quality inspection and evaluation form and the regulatory requirements of the water conservancy project; Determine the relationship between the semantic elements of the target clauses, use the semantic elements as nodes and the relationships between semantic elements as edges, and construct a graph using graph construction technology; Using data structuring technology, the graph is converted into structured approval decision factors.

7. The water conservancy project construction management information platform according to claim 1, characterized in that: The use of text classification algorithms to implement classified archiving management of quality inspection and evaluation forms includes: Using historical form data, we trained the support vector machine text classification algorithm module; Preprocess the data of the quality inspection form, and substitute the preprocessed form data into the support vector machine text classification algorithm module, and combine the preset classification model and rules to output the category of the quality inspection form; Through the archiving execution module, the quality inspection and evaluation forms are archived according to their categories.

8. The water conservancy project construction management information platform according to claim 1, characterized in that: The method of using the chaos test method and fuzzy cognitive map to predict the state evolution trend of water conservancy project risk factors includes: Obtain construction site time series data for water conservancy projects; perform chaos testing and phase space reconstruction on the construction site time series data, and build a local prediction model by optimizing prediction parameters to achieve short-term prediction of the time series of various risk factors of water conservancy projects; A fuzzy cognitive map of each risk factor node is constructed, and combined with the short-term prediction results of the time series of each risk factor of water conservancy project, the node status of each risk factor is updated, and the risk status matrix of water conservancy project construction in the future is generated.

9. The water conservancy project construction management information platform according to claim 8, characterized in that: The establishment of an early warning response mechanism and response plan includes: Based on the risk status matrix, combined with water conservancy project industry standards and historical experience, early warning thresholds are set, and risk levels of water conservancy project hazards are determined, and multi-level early warning information is output; Through plan matching rules and based on different risk levels and risk scenarios, corresponding dynamic plans are generated.

10. The water conservancy project construction management information platform according to claim 1, characterized in that: It also includes an intelligent analysis module for visualizing and personalizing water conservancy project construction data based on a big data platform.

Images