A digital management system for project information based on big data

Abstract

A project information digital management system based on big data relates to the field of digital management technology, and comprises a monitoring center, wherein the monitoring center is communicatively connected with a data acquisition module, a data analysis module, a data processing module, a data mining module, a model building module, a potential risk analysis module and a correction and early warning module; multi-source heterogeneous data are obtained, and progress status judgment and abnormality monitoring are performed on each engineering sub-stage in each project stage; the influence relationship between an internal feature data set and an external environment data set and development progress data is obtained; an external environment prediction model is constructed to obtain an estimated vulnerability coefficient of each type of feature data; the potential risk level of each project stage is obtained, and the collection time interval of each project stage is updated; the collection time interval of each project stage and the external environment prediction model are corrected, and early warning operations are performed on each project stage, thereby improving the execution efficiency and success rate of the project.

Description

Technical Field

[0001] The present invention relates to the field of digital management technology, and in particular to a project information digital management system based on big data. Background Art

[0002] The prior art CN115983808A "An intelligent management system and method for project data based on digital construction" includes capturing all approval records that have occurred in the digital project data management platform, extracting feature information from each approval record, and sorting out all approval events; sorting out the main approval condition element information that must be met for each type of approval event; judging and identifying approval events with approval associations; real-time monitoring and capturing of project data entry by project personnel in the digital project data management platform, and feeding back to the digital project data management platform to set corresponding priority rankings in approval events with approval associations, controlling the digital project data management platform to generate corresponding pending approval events based on priority rankings, and at the same time feeding back to relevant authorized users for event approval;

[0003] The prior art CN118172024A "A method and system for digital management of land engineering projects" obtains multimodal information data of land engineering projects and performs digital description analysis of dynamic processes based on the multimodal information data of land engineering projects to obtain digital description data of dynamic processes of land engineering projects; performs keyword extraction processing and project importance and complexity assessment analysis on the digital description data of dynamic processes of land engineering projects to obtain data on the importance of land engineering projects and data on the complexity of land engineering projects; performs project classification processing based on the data on the importance of land engineering projects and the data on the complexity of land engineering projects to obtain classification results of digital land engineering projects; performs differentiated management processing on the classification results of digital land engineering projects to obtain differentiated management results of classified land engineering projects.

[0004] During the project execution process, there are often changes in requirements, design adjustments, or changes in the market environment. These changes may affect the status of the project. Most engineering projects face complexity and uncertainty, including technological immaturity, market fluctuations, supply chain issues, etc. In order to help the management team effectively deal with various risks, some risks may be interrelated. Big data systems are usually early warnings for a single risk, and it may be difficult to capture the complex interactive effects between multiple risks, thereby affecting the overall risk control capabilities. Therefore, how to deal with this complexity and uncertainty, provide reliable analysis and prediction capabilities, and timely capture various changes in the project execution process and evaluate their impact, so as to timely adjust risk management strategies and countermeasures is an issue that we urgently need to solve. Summary of the invention

[0005] In order to solve the above technical problems, the object of the present invention is to provide a digital project information management system based on big data, including a monitoring center, wherein the monitoring center is communicatively connected with a data acquisition module, a data analysis module, a data processing module, a data mining module, a model building module, a potential risk analysis module and a correction warning module;

[0006] The data acquisition module and the monitoring center are connected in a distributed manner through the Internet of Things nodes to obtain multi-source heterogeneous data collected by several Internet of Things nodes, wherein the multi-source heterogeneous data includes development progress data, internal feature data sets, and external environment data sets;

[0007] The data analysis module is used to determine the progress status of each engineering sub-stage in each project stage according to the progress stage data of each project stage;

[0008] The data processing module is used to perform abnormal monitoring on various types of feature data in the internal feature data set of each project stage;

[0009] The data mining module is used to perform data mining on multi-source heterogeneous data in several historical collection cycles at each project stage to obtain the influence relationship between the internal feature data set and the external environment data set and development progress data;

[0010] The model building module is used to build an external environment prediction model to obtain the estimated numerical time series of various types of environmental data in the current collection period and the estimated vulnerability coefficients of various types of characteristic data in each project stage;

[0011] The potential risk analysis module is used to obtain the potential risk level of each project stage and update the collection time interval of each project stage;

[0012] The correction warning module is used to correct the collection time interval of each project stage and the external environment prediction model based on the multi-source heterogeneous data collected in real time, and perform early warning operations on each project stage.

[0013] Furthermore, the process of the data acquisition module acquiring multi-source heterogeneous data includes:

[0014] Acquire the project processing flow information of the project construction, extract the internal feature data set of each project according to the project processing flow information, divide the project construction into several project stages according to the internal feature data set of each project, set an Internet of Things node in each project stage, preset an initial collection time interval, and the Internet of Things node is used to collect multi-source heterogeneous data of each project stage at every initial collection time interval and transmit the multi-source heterogeneous data to the data analysis module, mark the collection time, and set the collection cycle.

[0015] Furthermore, the process of the data analysis module judging the progress status of each engineering sub-stage in each project stage according to the progress stage data of each project stage includes:

[0016] Obtain the functions of each engineering sub-stage in the project stage, and then divide the project stage into progress according to the functions of each engineering sub-stage. One function is regarded as a stage to form a progress stage. Then, estimate the development time of the progress stage, obtain the estimated time of the progress stage, bind the estimated time with the progress stage, and form the completion progress monitoring items of each progress stage.

[0017] The end timestamp of each engineering sub-stage is used as a progress detection point, and the actual development time of the progress stage is collected at the progress detection point. When the actual development time of the progress stage is greater than the estimated time of the progress stage, the timeout time of the engineering sub-stage corresponding to the progress stage is obtained according to the actual development time and the estimated time, and the engineering sub-stage corresponding to the progress stage is marked as an abnormal progress state. When the actual development time of the progress stage is less than or equal to the estimated time of the progress stage, the engineering sub-stage corresponding to the progress stage is marked as a normal progress state.

[0018] Furthermore, the process of the data processing module performing abnormal monitoring on each type of feature data in the internal feature data set includes:

[0019] Preset the threshold interval of each type of feature data in the internal feature data set of each project stage, compare the value of each type of feature data in the internal feature data set collected by each IoT node in the current collection period with the corresponding threshold interval, and if there is feature type data whose value is not within the corresponding threshold interval, generate an abnormal alarm signal for the feature type data.

[0020] Furthermore, the data mining module performs data mining on multi-source heterogeneous data in several historical collection cycles at each project stage, and the process of obtaining the influence relationship between the internal feature data set and the external environment data set and the development progress data includes:

[0021] Acquire development progress data, internal feature data sets, and external environment data sets in several historical collection cycles for each project stage, extract numerical time series sequences of each type of feature data in the development progress data and the internal feature data sets, and numerical time series sequences of each type of environmental data in the external environment data sets for statistical analysis, obtain the numerical range interval of the numerical time series sequence of each type of environmental data, select a threshold point in the numerical range interval, divide the numerical range interval into different fuzzy sub-intervals, obtain the probability of each type of feature data generating an abnormal alarm signal in different fuzzy sub-intervals of each type of environmental data and the fluctuation coefficient of each type of feature data, and obtain the vulnerability coefficient corresponding to each type of feature data in each project stage in different fuzzy sub-intervals of each type of environmental data according to the probability and the fluctuation coefficient;

[0022] When obtaining each type of feature data in the internal feature data set of each project stage in several historical collection cycles to generate an abnormal alarm signal, the probability of each engineering sub-stage of each project stage being in an abnormal progress state and the average timeout time corresponding to the abnormal state of each engineering sub-stage of each project stage are obtained, and the progress stage influence coefficient of each type of feature data in each project stage on each engineering sub-stage of each project stage is obtained according to the probability and the average timeout time.

[0023] Furthermore, the model building module builds an external environment prediction model, and the process of obtaining the estimated numerical time series of various types of environmental data in the current collection period and the estimated vulnerability coefficients of various types of characteristic data in each project stage includes:

[0024] An external environment prediction model is constructed based on deep learning, and various types of environmental data within several historical collection cycles within a preset range of the construction project are obtained through GIS as a test set and a training set; the training set is input into the external environment prediction model for training until the loss function training is stable, and the model parameters are saved, and the external environment prediction model is tested through the test set until it meets the preset requirements, and the external environment prediction model is output;

[0025] According to the trained external environment prediction model, the estimated numerical time series sequence of each type of environmental data in the current collection cycle is obtained, and the fuzzy sub-interval to which the estimated numerical time series sequence of each type of environmental data belongs is obtained. According to the fuzzy sub-interval and the corresponding vulnerability coefficients of each type of feature data in different fuzzy sub-intervals of each type of environmental data, the estimated vulnerability coefficients of each type of feature data in each project stage in the current collection cycle are obtained.

[0026] Further, the potential risk analysis module obtains the potential risk level of each project stage according to the estimated vulnerability coefficient of each type of characteristic data of each project stage and the progress phase impact coefficient of each type of characteristic data on each engineering sub-stage of the project stage, and updates the collection time interval of each project stage. The process includes:

[0027] The influence coefficient of each type of characteristic data in the project stage on the progress stage of each engineering sub-stage of the project stage is taken as the evaluation index, and the estimated vulnerability coefficient of each type of characteristic data in the current collection cycle is taken as the weight coefficient. The indicator weight matrix of the influence coefficient of each type of characteristic data on the progress stage of each engineering sub-stage is constructed, the potential risk level is preset, and the membership matrix of the project stage to the potential risk level is obtained through fuzzy comprehensive evaluation;

[0028] The potential risk level of the project stage is obtained through the membership matrix and the indicator weight matrix, and a potential risk level threshold is preset. When the potential risk level of the project stage is greater than the potential risk level threshold, the project construction of the project stage is suspended. When the potential risk level of the project stage is less than or equal to the potential risk level threshold, a collection time interval comparison table is constructed. The collection time interval table includes collection time intervals corresponding to different potential risk levels. The collection time interval of the project stage is generated according to the potential risk level of the project stage and the collection time interval comparison table, and the initial collection time interval of the Internet of Things node corresponding to the project stage of the current collection cycle is updated to the collection time interval.

[0029] Furthermore, the correction warning module corrects the collection time interval of each project stage and the external environment prediction model according to the multi-source heterogeneous data collected in real time, and performs early warning operations on each project stage, including:

[0030] Acquire the development progress data of the project stage in the current collection cycle, the real-time numerical time series sequence of each type of feature data in the internal feature data set, and the real-time numerical time series sequence of each type of environmental data in the external environment data set, compare the real-time numerical time series sequence of each type of environmental data with the corresponding estimated numerical time series sequence, obtain the similarity between the real-time numerical time series sequence and the estimated numerical time series sequence, preset a similarity threshold, and if the similarity is less than the similarity threshold, obtain the collection time interval of the project stage based on the real-time numerical time series sequence of each type of environmental data, use the collection time interval as the new collection time interval for the Internet of Things nodes in the project stage, and include the real-time numerical time series sequence of each type of environmental data in the test set and the training set to re-train the external environment prediction model;

[0031] At the same time, a time proportion threshold and a fluctuation coefficient threshold are preset, and the time proportion of the actual development time and the expected time of each engineering sub-stage within the project stage is obtained according to the development progress data. When the time proportion of each engineering sub-stage is equal to the time proportion threshold, the fluctuation coefficient of each type of feature data is obtained according to the real-time numerical time series sequence of each type of feature data, and the fluctuation coefficient is compared with the fluctuation coefficient threshold. If there is a type of feature data with a fluctuation coefficient greater than the fluctuation coefficient threshold, an early warning signal of the type of feature data is generated.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] 1. Judge the progress status of each engineering sub-stage in each project stage according to the progress stage data of each project stage, monitor the abnormality of various types of feature data in the internal feature data of each project stage, and collect and monitor the data on the construction site in real time, including construction progress, resource utilization (such as manpower, materials), equipment status, etc., to provide comprehensive real-time monitoring, accurate decision support and effective resource optimization for construction and engineering project management, thereby improving the execution efficiency and success rate of the project.

[0034] 2. Conduct data mining on multi-source heterogeneous data in several historical collection cycles for each project stage, obtain the influence relationship between the internal feature data set and the external environment data set and development progress data, build an external environment prediction model, obtain the estimated numerical time series of each type of environmental data in the current collection cycle and the estimated vulnerability coefficient of each type of feature data in each project stage, obtain the potential risk level of each project stage, identify potential risk factors in the project by analyzing historical data and real-time monitoring results, such as construction delays, cost overruns, quality problems, etc., and update the collection time interval for each project stage. Correct the collection time interval for each project stage and the external environment prediction model based on the multi-source heterogeneous data collected in real time, perform early warning operations for each project stage, and through continuous accumulation and analysis of data, help the team to timely warn and respond to various potential risks during project execution, improve the success rate of the project and overall management efficiency, and achieve continuous improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a schematic diagram of a project information digital management system based on big data according to an embodiment of the present application. DETAILED DESCRIPTION

[0036] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0037] like Figure 1 As shown, a digital management system for project information based on big data includes a monitoring center, which is communicatively connected to a data acquisition module, a data analysis module, a data processing module, a data mining module, a model building module, a potential risk analysis module and a correction warning module;

[0038] The data acquisition module and the monitoring center are connected in a distributed manner through the Internet of Things nodes to obtain multi-source heterogeneous data collected by several Internet of Things nodes, wherein the multi-source heterogeneous data includes development progress data, internal feature data sets, and external environment data sets;

[0039] The data analysis module is used to determine the progress status of each engineering sub-stage in each project stage according to the progress stage data of each project stage;

[0040] The data processing module is used to perform abnormal monitoring on various types of feature data in the internal feature data set of each project stage;

[0041] The data mining module is used to perform data mining on multi-source heterogeneous data in several historical collection cycles at each project stage to obtain the influence relationship between the internal feature data set and the external environment data set and development progress data;

[0042] The model building module is used to build an external environment prediction model to obtain the estimated numerical time series of various types of environmental data in the current collection period and the estimated vulnerability coefficients of various types of characteristic data in each project stage;

[0043] The potential risk analysis module is used to obtain the potential risk level of each project stage and update the collection time interval of each project stage;

[0044] The correction warning module is used to correct the collection time interval of each project stage and the external environment prediction model based on the multi-source heterogeneous data collected in real time, and perform early warning operations on each project stage.

[0045] It should be further explained that, in the specific implementation process, the process of the data acquisition module acquiring multi-source heterogeneous data includes:

[0046] Acquire the project processing flow information of the project construction, extract the internal feature data set of each project according to the project processing flow information, divide the project construction into several project stages according to the internal feature data set of each project, set an Internet of Things node in each project stage, preset an initial collection time interval, and the Internet of Things node is used to collect multi-source heterogeneous data of each project stage at every initial collection time interval and transmit the multi-source heterogeneous data to the data analysis module, mark the collection time, and set the collection cycle.

[0047] It should be further explained that, in the specific implementation process, the project phase of project construction and the various engineering sub-phases in the project phase include:

[0048] Design phase:

[0049] Preliminary design sub-stage: preliminary formulation of the overall design plan of the project, including building structure, system layout, etc.; Construction drawing design sub-stage: according to the preliminary design plan, detailed construction drawings are formulated, including drawings of architecture, structure, water supply and drainage, electricity, etc.;

[0050] Bidding and procurement stage:

[0051] Tender preparation sub-stage: prepare tender documents, clarify project requirements, contract terms, bid evaluation criteria, etc. Contractor selection sub-stage: issue tender notices, review bid documents and select suitable contractors or suppliers;

[0052] Construction phase:

[0053] Construction preparation sub-phase: determine the construction plan and schedule, prepare the construction site and necessary facilities; construction implementation sub-phase: perform construction work, including civil construction, equipment installation, system commissioning, etc.; quality control sub-phase: implement quality management measures to ensure that the construction quality meets the design requirements and standards;

[0054] Completion and delivery phase:

[0055] Acceptance and testing sub-phase: complete various tests and trial runs of the project; Completion acceptance sub-phase: conduct final acceptance, confirm that the project has been completed in accordance with the contract requirements, and complete the handover procedures.

[0056] It should be further explained that, in the specific implementation process, the various types of feature data in the internal feature data set include:

[0057] Resource consumption data: including material usage, material costs, labor costs, and equipment rental costs for each engineering sub-stage during the project phase;

[0058] Equipment status data: including equipment operation status data of each engineering sub-stage in the project phase, such as equipment operation time, equipment working status (such as running, stopped, maintenance), equipment energy consumption, such as the use of electricity, fuel, water and other resources;

[0059] Quality inspection data: including construction quality inspection results, defect repair status, acceptance pass rate, etc.

[0060] It should be further explained that, in the specific implementation process, the various types of environmental data in the external environment data set include:

[0061] Meteorological data: including temperature, humidity, precipitation, wind speed and direction, which have an important impact on construction planning, material selection and safety management;

[0062] Geological and topographic data: including soil type, geological structure, groundwater level and terrain relief, which are crucial for basic engineering design, foundation treatment and construction safety;

[0063] Environmental protection data: including data on air quality, water quality monitoring, noise levels, and surrounding environmental impact assessments. These data must comply with environmental protection laws and regulations and guide the project's environmental management and social responsibility.

[0064] Traffic and logistics data: including data on road conditions, traffic flow, transportation networks and logistics channels. These data are of great significance for the transportation arrangement of construction materials, traffic management during construction, and project progress control;

[0065] Supply chain risk data: including data such as material supply stability, supplier credibility, and logistics and transportation conditions, which may affect project construction progress and cost control.

[0066] It should be further explained that, in the specific implementation process, the process of the data analysis module judging the progress status of each engineering sub-stage in each project stage according to the progress stage data of each project stage includes:

[0067] Obtain the functions of each engineering sub-stage in the project stage, and then divide the project stage into progress according to the functions of each engineering sub-stage. One function is regarded as a stage to form a progress stage. Then, estimate the development time of the progress stage, obtain the estimated time of the progress stage, bind the estimated time with the progress stage, and form the completion progress monitoring items of each progress stage.

[0068] The end timestamp of each engineering sub-stage is used as a progress detection point, and the actual development time of the progress stage is collected at the progress detection point. When the actual development time of the progress stage is greater than the estimated time of the progress stage, the timeout time of the engineering sub-stage corresponding to the progress stage is obtained according to the actual development time and the estimated time, and the engineering sub-stage corresponding to the progress stage is marked as an abnormal progress state. When the actual development time of the progress stage is less than or equal to the estimated time of the progress stage, the engineering sub-stage corresponding to the progress stage is marked as a normal progress state.

[0069] It should be further explained that, in the specific implementation process, the process of the data processing module performing abnormal monitoring on each type of feature data in the internal feature data set includes:

[0070] Preset the threshold interval of each type of feature data in the internal feature data set of each project stage, compare the value of each type of feature data in the internal feature data set collected by each IoT node in the current collection period with the corresponding threshold interval, and if there is feature type data whose value is not within the corresponding threshold interval, generate an abnormal alarm signal for the feature type data.

[0071] It should be further explained that, in the specific implementation process, the data mining module performs data mining on multi-source heterogeneous data in several historical collection cycles at each project stage, and the process of obtaining the influence relationship between the internal feature data set and the external environment data set and the development progress data includes:

[0072] The development progress data, internal feature data sets and external environment data sets of each project stage in several historical collection cycles are obtained, the numerical time series sequences of each type of feature data in the development progress data and the internal feature data sets and the numerical time series sequences of each type of environmental data in the external environment data sets are extracted for statistical analysis, the numerical range interval of the numerical time series sequence of each type of environmental data is obtained, the threshold point is selected in the numerical range interval, the numerical range interval is divided into different fuzzy sub-intervals, the probability of each type of feature data generating an abnormal alarm signal and the fluctuation coefficient of each type of feature data in different fuzzy sub-intervals of each type of environmental data are obtained, and the vulnerability coefficient corresponding to each type of feature data in each project stage in different fuzzy sub-intervals of each type of environmental data is obtained according to the probability and the fluctuation coefficient, wherein the calculation formula for obtaining the vulnerability coefficient corresponding to each type of feature data in each project stage in different fuzzy sub-intervals of each type of environmental data according to the probability and the fluctuation coefficient is:

[0073] tsi = δ × a1pd × a2dp;

[0074] Among them, tsi represents the vulnerability coefficient, pd represents the probability, dp represents the volatility coefficient, a1 and a2 represent the weight factors, and δ represents the conversion coefficient;

[0075] Obtain the probability of each engineering sub-stage of each project stage being in an abnormal progress state and the average timeout time corresponding to each engineering sub-stage of each project stage being in an abnormal state when each type of feature data in the internal feature data set in several historical collection cycles generates an abnormal alarm signal, and obtain the progress stage influence coefficient of each type of feature data in each project stage for each engineering sub-stage of each project stage according to the probability and the average timeout time, wherein the process of obtaining the progress stage influence coefficient of each type of feature data in each project stage for each engineering sub-stage of each project stage according to the probability and the average timeout time is:

[0076] jkl = θ × a3ps × a4ts;

[0077] Among them, jkl represents the progress stage impact coefficient, ps represents the probability, ts represents the average overtime time, a3 and a4 represent weight factors, and θ represents the conversion coefficient.

[0078] It should be further explained that, in the specific implementation process, the model building module builds an external environment prediction model, and the process of obtaining the estimated numerical time series of various types of environmental data in the current collection period and the estimated vulnerability coefficients of various types of characteristic data in each project stage includes:

[0079] An external environment prediction model is constructed based on deep learning, and various types of environmental data within several historical collection cycles within a preset range of the construction project are obtained through GIS as a test set and a training set; the training set is input into the external environment prediction model for training until the loss function training is stable, and the model parameters are saved, and the external environment prediction model is tested through the test set until it meets the preset requirements, and the external environment prediction model is output;

[0080] According to the trained external environment prediction model, the estimated numerical time series sequence of each type of environmental data in the current collection cycle is obtained, and the fuzzy sub-interval to which the estimated numerical time series sequence of each type of environmental data belongs is obtained. According to the fuzzy sub-interval and the corresponding vulnerability coefficients of each type of feature data in different fuzzy sub-intervals of each type of environmental data, the estimated vulnerability coefficients of each type of feature data in each project stage in the current collection cycle are obtained.

[0081] It should be further explained that, in the specific implementation process, the potential risk analysis module obtains the potential risk level of each project stage according to the estimated vulnerability coefficient of each type of characteristic data of each project stage and the progress phase influence coefficient of each type of characteristic data on each engineering sub-stage of the project stage, and updates the collection time interval of each project stage. The process includes:

[0082] The influence coefficient of each type of characteristic data in the project stage on the progress stage of each engineering sub-stage of the project stage is taken as the evaluation index, and the estimated vulnerability coefficient of each type of characteristic data in the current collection cycle is taken as the weight coefficient. The indicator weight matrix of the influence coefficient of each type of characteristic data on the progress stage of each engineering sub-stage is constructed, the potential risk level is preset, and the membership matrix of the project stage to the potential risk level is obtained through fuzzy comprehensive evaluation;

[0083] The potential risk level of the project stage is obtained through the membership matrix and the indicator weight matrix, and a potential risk level threshold is preset. When the potential risk level of the project stage is greater than the potential risk level threshold, the project construction of the project stage is suspended. When the potential risk level of the project stage is less than or equal to the potential risk level threshold, a collection time interval comparison table is constructed. The collection time interval table includes collection time intervals corresponding to different potential risk levels. The collection time interval of the project stage is generated according to the potential risk level of the project stage and the collection time interval comparison table, and the initial collection time interval of the Internet of Things node corresponding to the project stage of the current collection cycle is updated to the collection time interval.

[0084] It should be further explained that, in the specific implementation process, the process of obtaining the potential risk level of each project stage through the membership matrix and the indicator weight matrix includes:

[0085] The indicator weight matrix and the membership matrix of the evaluation indicators are integrated through a formula to obtain a fuzzy comprehensive evaluation matrix of the evaluation indicators, and the membership of each project stage to different potential risk levels is obtained according to the fuzzy comprehensive evaluation matrix, and the potential risk level with the highest membership corresponding to each project stage is screened out, and the potential risk level with the highest membership corresponding to each project stage is used as the potential risk level of each project stage;

[0086] Wherein, the formula is:

[0087] M = αM1 × βM2;

[0088] Among them, M is the fuzzy comprehensive evaluation matrix of the evaluation index, M1 is the indicator weight matrix of the evaluation index, M2 is the membership matrix, "×" represents the multiplication of the elements at the corresponding positions of the weight matrix of the evaluation index and the membership matrix, and α and β are weighting parameters for controlling the balance between the weight matrix and the membership matrix in the fuzzy comprehensive evaluation matrix of the evaluation index.

[0089] It should be further explained that, in the specific implementation process, the correction warning module corrects the collection time interval of each project stage and the external environment prediction model based on the multi-source heterogeneous data collected in real time, and performs early warning operations for each project stage, including:

[0090] Acquire the development progress data of the project stage in the current collection cycle, the real-time numerical time series sequence of each type of feature data in the internal feature data set, and the real-time numerical time series sequence of each type of environmental data in the external environment data set, compare the real-time numerical time series sequence of each type of environmental data with the corresponding estimated numerical time series sequence, obtain the similarity between the real-time numerical time series sequence and the estimated numerical time series sequence, preset a similarity threshold, and if the similarity is less than the similarity threshold, obtain the collection time interval of the project stage based on the real-time numerical time series sequence of each type of environmental data, use the collection time interval as the new collection time interval for the Internet of Things nodes in the project stage, and include the real-time numerical time series sequence of each type of environmental data in the test set and the training set to re-train the external environment prediction model;

[0091] At the same time, a time proportion threshold and a fluctuation coefficient threshold are preset, and the time proportion of the actual development time and the expected time of each engineering sub-stage within the project stage is obtained according to the development progress data. When the time proportion of each engineering sub-stage is equal to the time proportion threshold, the fluctuation coefficient of each type of feature data is obtained according to the real-time numerical time series sequence of each type of feature data, and the fluctuation coefficient is compared with the fluctuation coefficient threshold. If there is a type of feature data with a fluctuation coefficient greater than the fluctuation coefficient threshold, an early warning signal of the type of feature data is generated.

[0092] It should be further explained that, in the specific implementation process, the real-time numerical time series sequence of the type environment data is compared with the corresponding estimated numerical time series sequence, and the calculation formula for obtaining the similarity between the real-time numerical time series sequence and the estimated numerical time series sequence and the fluctuation coefficient is:

[0093]

[0094] Among them, zs represents the similarity, dp represents the fluctuation coefficient, and D itrepresents the value of the real-time numerical time series at the tth moment; DiA represents the average value of the real-time numerical time series; D jt represents the value of the estimated numerical time series at the tth moment; DjA represents the average value of the estimated numerical time series; n represents the total number of moments.

[0095] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A digital project information management system based on big data, characterized in that: It includes a monitoring center, which is communicatively connected with a data acquisition module, a data analysis module, a data processing module, a data mining module, a model building module, a potential risk analysis module and a correction warning module; The data acquisition module and the monitoring center are connected in a distributed manner through the Internet of Things nodes to obtain multi-source heterogeneous data collected by several Internet of Things nodes, wherein the multi-source heterogeneous data includes development progress data, internal feature data sets, and external environment data sets; The data analysis module is used to determine the progress status of each engineering sub-stage in each project stage according to the progress stage data of each project stage; The data processing module is used to perform abnormal monitoring on various types of feature data in the internal feature data set of each project stage; The data mining module is used to perform data mining on multi-source heterogeneous data in several historical collection cycles at each project stage to obtain the influence relationship between the internal feature data set and the external environment data set and development progress data, including: Acquire development progress data, internal feature data sets, and external environment data sets in several historical collection cycles for each project stage, extract numerical time series sequences of each type of feature data in the development progress data and the internal feature data sets, and numerical time series sequences of each type of environmental data in the external environment data sets for statistical analysis, obtain the numerical range interval of the numerical time series sequence of each type of environmental data, select a threshold point in the numerical range interval, divide the numerical range interval into different fuzzy sub-intervals, obtain the probability of each type of feature data generating an abnormal alarm signal in different fuzzy sub-intervals of each type of environmental data and the fluctuation coefficient of each type of feature data, and obtain the vulnerability coefficient corresponding to each type of feature data in each project stage in different fuzzy sub-intervals of each type of environmental data according to the probability and the fluctuation coefficient; The calculation formula for obtaining the vulnerability coefficient corresponding to each type of characteristic data in each project stage in different fuzzy sub-intervals of each type of environmental data according to the probability and fluctuation coefficient is as follows: ； Among them, tsi represents the vulnerability coefficient, pd represents the probability, dp represents the volatility coefficient, a1 and a2 represent the weight factors, Indicates the conversion factor; Obtaining the probability of each engineering sub-stage of each project stage being in an abnormal progress state and the average timeout time corresponding to each engineering sub-stage of each project stage being in an abnormal state when each type of feature data in the internal feature data set in several historical collection cycles generates an abnormal alarm signal, and obtaining the progress stage influence coefficient of each type of feature data in each project stage on each engineering sub-stage of each project stage according to the probability and the average timeout time; The calculation formula for obtaining the progress phase influence coefficient of each type of characteristic data in each project phase for each engineering sub-phase of each project phase according to the probability and the average overtime time is as follows: ts; in, represents the progress stage impact coefficient, represents probability, ts represents average timeout time, a3 and a4 represent weight factors, Indicates the conversion factor; The model building module is used to build an external environment prediction model to obtain the estimated numerical time series of various types of environmental data in the current collection period and the estimated vulnerability coefficients of various types of characteristic data in each project stage, including: An external environment prediction model is constructed based on deep learning, and various types of environmental data within several historical collection cycles within a preset range of the construction project are obtained through GIS as a test set and a training set; the training set is input into the external environment prediction model for training until the loss function training is stable, and the model parameters are saved, and the external environment prediction model is tested through the test set until it meets the preset requirements, and the external environment prediction model is output; Obtain an estimated numerical time series sequence of each type of environmental data in the current collection cycle according to the trained external environment prediction model, obtain the fuzzy sub-interval to which the estimated numerical time series sequence of each type of environmental data belongs, and obtain the estimated vulnerability coefficient of each type of characteristic data in the current collection cycle at each project stage according to the fuzzy sub-interval and the corresponding vulnerability coefficient of each type of characteristic data in different fuzzy sub-intervals of each type of environmental data; The potential risk analysis module is used to obtain the potential risk level of each project stage and update the collection time interval of each project stage, including: The influence coefficient of each type of characteristic data in the project stage on the progress stage of each engineering sub-stage of the project stage is taken as the evaluation index, and the estimated vulnerability coefficient of each type of characteristic data in the current collection cycle is taken as the weight coefficient. The indicator weight matrix of the influence coefficient of each type of characteristic data on the progress stage of each engineering sub-stage is constructed, the potential risk level is preset, and the membership matrix of the project stage to the potential risk level is obtained through fuzzy comprehensive evaluation; The potential risk level of the project stage is obtained through the membership matrix and the indicator weight matrix, and a potential risk level threshold is preset. When the potential risk level of the project stage is greater than the potential risk level threshold, the project construction of the project stage is suspended. When the potential risk level of the project stage is less than or equal to the potential risk level threshold, a collection time interval comparison table is constructed, and the collection time interval table includes collection time intervals corresponding to different potential risk levels. The collection time interval of the project stage is generated according to the potential risk level of the project stage and the collection time interval comparison table, and the initial collection time interval of the Internet of Things node corresponding to the current collection cycle project stage is updated to the collection time interval; The correction warning module is used to correct the collection time interval of each project stage and the external environment prediction model based on the multi-source heterogeneous data collected in real time, and perform early warning operations on each project stage, including: Acquire the development progress data of the project stage in the current collection cycle, the real-time numerical time series sequence of each type of feature data in the internal feature data set, and the real-time numerical time series sequence of each type of environmental data in the external environment data set, compare the real-time numerical time series sequence of each type of environmental data with the corresponding estimated numerical time series sequence, obtain the similarity between the real-time numerical time series sequence and the estimated numerical time series sequence, preset a similarity threshold, and if the similarity is less than the similarity threshold, obtain the collection time interval of the project stage based on the real-time numerical time series sequence of each type of environmental data, use the collection time interval as the new collection time interval for the Internet of Things nodes in the project stage, and include the real-time numerical time series sequence of each type of environmental data in the test set and the training set to re-train the external environment prediction model; At the same time, a time ratio threshold and a fluctuation coefficient threshold are preset, and the time ratio of the actual development time and the expected time of each engineering sub-stage in the project stage is obtained according to the development progress data. When the time ratio of each engineering sub-stage is equal to the time ratio threshold, the fluctuation coefficient of each type of feature data is obtained according to the real-time numerical time series sequence of each type of feature data, and the fluctuation coefficient is compared with the fluctuation coefficient threshold. If there is a type of feature data with a fluctuation coefficient greater than the fluctuation coefficient threshold, an early warning signal of the type of feature data is generated; Compare the real-time numerical time series of type environment data with the corresponding estimated numerical time series, and obtain the similarity between the real-time numerical time series and the estimated numerical time series, as well as the calculation formula of the fluctuation coefficient: ； ； Among them, zs represents similarity, dp represents fluctuation coefficient, Represents the value of the real-time numerical time series at the tth moment; Represents the average value of a real-time numerical time series; Represents the value of the estimated numerical time series at the tth moment; Represents the average value of the estimated numerical time series; n represents the total number of moments.

2. According to the big data-based project information digital management system of claim 1, it is characterized in that: The process of the data acquisition module acquiring multi-source heterogeneous data includes: Acquire the project processing flow information of the project construction, extract the internal feature data set of each project according to the project processing flow information, divide the project construction into several project stages according to the internal feature data set of each project, set an Internet of Things node in each project stage, preset an initial collection time interval, and the Internet of Things node is used to collect multi-source heterogeneous data of each project stage at every initial collection time interval and transmit the multi-source heterogeneous data to the data analysis module, mark the collection time, and set the collection cycle.

3. A digital project information management system based on big data according to claim 2, characterized in that: The process of the data analysis module judging the progress status of each engineering sub-stage in each project stage according to the progress stage data of each project stage includes: Obtain the functions of each engineering sub-stage in the project stage, and then divide the project stage into progress according to the functions of each engineering sub-stage. One function is regarded as a stage to form a progress stage. Then, estimate the development time of the progress stage, obtain the estimated time of the progress stage, bind the estimated time with the progress stage, and form the completion progress monitoring items of each progress stage. The end timestamp of each engineering sub-stage is used as a progress detection point, and the actual development time of the progress stage is collected at the progress detection point. When the actual development time of the progress stage is greater than the estimated time of the progress stage, the timeout time of the engineering sub-stage corresponding to the progress stage is obtained according to the actual development time and the estimated time, and the engineering sub-stage corresponding to the progress stage is marked as an abnormal progress state. When the actual development time of the progress stage is less than or equal to the estimated time of the progress stage, the engineering sub-stage corresponding to the progress stage is marked as a normal progress state.

4. A digital project information management system based on big data according to claim 3, characterized in that: The process of the data processing module performing abnormal monitoring on each type of feature data in the internal feature data set includes: Preset the threshold interval of each type of feature data in the internal feature data set of each project stage, compare the value of each type of feature data in the internal feature data set collected by each IoT node in the current collection period with the corresponding threshold interval, and if there is feature type data whose value is not within the corresponding threshold interval, generate an abnormal alarm signal for the feature type data.

Images