Engineering quality safety digital management method integrated with large language model

Abstract

The invention provides an engineering quality safety digital management method integrated with a large language model, and the method comprises the following steps: S1, extracting engineering feature data from a basic database, and building an engineering quality safety knowledge graph according to the engineering feature data, the engineering feature data including process feature data, material feature data and safety monitoring data; s2, acquiring engineering field data of a current engineering project; and S3, according to the established engineering quality safety knowledge graph, performing engineering quality safety analysis on the engineering field data based on the knowledge graph enhanced large language model, and generating an engineering quality safety evaluation result. According to the method, the pertinence and the application effect of the large language model on quality safety management and control of the engineering project can be improved, and the intelligent level of digital management of the engineering project is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of construction engineering quality and safety management, and in particular to a digital management method for engineering quality and safety that incorporates a large language model. Background Art

[0002] The quality and safety of construction projects are directly related to the safety of life and property of the nation and its people, as well as social stability. However, faced with complex engineering environments, diverse stakeholders, and multimodal data and information, traditional construction quality and safety management models have gradually exposed problems such as inefficient digital management, poor information flow, and insufficient decision-making support.

[0003] Artificial intelligence technologies based on deep learning, image recognition, and natural language processing are key to digital management of construction project quality and safety and are gaining widespread adoption and recognition in construction project management. Large language model technology is also playing a crucial role in digital management of enterprise project quality and safety.

[0004] However, in the current process of applying general large language models to solve the digital management of engineering quality and safety in enterprises, there are still problems such as insufficient intelligence and lack of pertinence, which leads to poor use results. Summary of the Invention

[0005] In response to the above problems, the present invention aims to provide a digital management method for engineering quality and safety that incorporates a large language model.

[0006] The purpose of the present invention is achieved by adopting the following technical solutions: The present invention proposes a digital management method for engineering quality and safety that incorporates a large language model, comprising the following steps: S1 extracts engineering feature data from the basic database and builds an engineering quality and safety knowledge graph based on the engineering feature data, where the engineering feature data includes process feature data, material feature data, and safety monitoring data; S2 obtains the engineering site data of the current engineering project; Based on the established engineering quality and safety knowledge graph, S3 conducts engineering quality and safety analysis on engineering site data using the knowledge graph-enhanced large language model and generates engineering quality and safety evaluation results.

[0007] Preferably, step S1 includes: S11 obtains construction data of historical projects from a basic database, and extracts process characteristic data based on the construction data, wherein the process characteristic data includes process name, construction period information, and process scheduling information; S12 further extracts material characteristic data from the construction data, wherein the material characteristic data includes material type, material performance record data, corresponding process and usage location, etc.; S13 extracts safety monitoring data based on the engineering records of historical projects, where the safety monitoring data includes recorded quality safety events, processes corresponding to the safety events, and rules and clauses associated with the safety events; S14 builds an engineering quality and safety knowledge graph based on the acquired process characteristic data, material characteristic data and safety monitoring data, and completes the setting of experience parameters in the engineering quality and safety knowledge graph.

[0008] Preferably, the engineering quality and safety knowledge graph constructed includes three types of nodes: process nodes, material nodes, and safety nodes; each process node corresponds to a process, each material node corresponds to a material, and each safety node corresponds to a type of quality and safety event; During the construction phase: Based on the process characteristic data of historical projects, obtain the construction period information of each process node, and calculate the delay probability of each process node based on the construction period time information ,in represents the delay probability of process i, Represents the statistical number of process i in the historical project, represents the number of delays of process i in the historical project; According to the material characteristic data of historical projects, the performance attenuation function corresponding to each material node is calculated ,in represents the performance decay function of material k at time t, represents the initial performance value of material k, represents the decay rate parameter, Indicates the ambient temperature, Indicates the ambient humidity, Represents the protective measure parameters, represents the environmental sensitivity parameter of material k; Based on the safety monitoring data of historical projects, the probability of quality safety incidents corresponding to each safety node causing abnormalities in the process or materials is calculated ,in Indicates a security incident The probability of causing an exception, Indicates security events in historical projects The total number of times recorded, Indicates security events in historical projects The number of times an anomaly occurs in the associated process or material while being recorded.

[0009] Preferably, the engineering site data of the current engineering project includes construction plan information, construction log, on-site monitoring data and safety monitoring data; wherein step S2 includes: S21 acquires corresponding planned process characteristic data based on the construction plan information of the current project, wherein the planned process characteristic data includes process name, planned construction period information, planned process scheduling information, and planned process material information, and further acquires actual process characteristic data based on the construction log of the current project, wherein the actual process characteristic data includes process name, actual construction period information, actual process scheduling information, and actual process material consumption information, etc.; S22 obtains material performance parameters and on-site environmental data based on on-site monitoring data of the current project, wherein the material performance parameters are obtained through material test reports, and the on-site environmental data are obtained through sensors installed at the construction site; S23 obtains recorded quality and safety event information based on the safety monitoring data of the current engineering project.

[0010] Preferably, step S3 includes: S31 updates the engineering site data to the engineering quality and safety knowledge graph based on the acquired engineering site data; S32 performs engineering quality and safety analysis based on the updated safety knowledge graph and the knowledge graph-enhanced large language model to obtain the engineering quality and safety analysis results; S33 performs transformation based on the obtained engineering quality and safety analysis results to obtain a visualized engineering quality and safety evaluation result.

[0011] Preferably, in step S31, updating the project site data to the project quality and safety knowledge graph specifically includes: In the initialization phase: update the corresponding process nodes according to the obtained planned process feature data, activate and connect the corresponding process nodes to form a process chain, where each process node contains the planned start time and planned end time of the process, the related process node information and the required material information; adjust the association weight between the process node and the corresponding material node according to the required material information ,in Indicates process node With material nodes The association weight between ; In the dynamic update stage: update the process nodes according to the actual process feature data obtained, update the process status and actual duration information to the corresponding process nodes; further obtain the dependency strength between the associated process nodes in the process chain ,in Indicates the dependence strength of process node j on process node i; Represents the delay probability of process node i. When process node i is delayed compared with the planned end time, ,otherwise , It represents the minimum waiting time of process node j after process node i is completed; represents the time length between the planned / actual completion time of process node i and the planned start time of process node j; Activate and update the material nodes based on the obtained on-site monitoring data, update the material performance parameters and on-site environmental data to the material nodes, and further estimate the real-time material performance attenuation value ,in Represents the material performance attenuation value of material k after time t, where t represents the time difference between the current moment and the initial performance value recorded by the material, Indicates the initial performance value of material k recorded, obtained according to the test report, Indicates the ambient temperature, Indicates the ambient humidity, which is obtained based on the data collected by the environmental sensor. Represents the protective measure parameter, obtained according to the current protective measures for material k, represents the environmental sensitivity parameter of material k; Update the safety node based on the obtained safety monitoring data, activate the corresponding safety node based on the recorded quality safety event, and record the quality safety event information in the safety node, where the quality safety event information includes the type of safety event, the time of occurrence, the corresponding material or process information, and the associated regulatory provisions; further calculate the safety triggering strength of the safety node ,in Indicates a security incident The probability of causing an exception, Indicates the process currently in progress Security incident The probability of triggering Indicates activation parameters. When security events are recorded, otherwise .

[0012] Preferably, in step S32, the engineering quality and safety analysis is performed based on the knowledge graph-enhanced large language model, including: S321 detects the dependency strength between the current process nodes, compares the dependency strength with the set dependency standard value, and obtains the delay risk analysis result; S322 detects the material performance attenuation value of the current material node, compares the material performance attenuation value with the set material performance standard value, and obtains the material performance analysis result; S323 detects the security triggering strength of the current security node, compares the security triggering strength with the set security standard value, and obtains the security event analysis result; Evolution is performed based on the current engineering quality and safety knowledge graph and the engineering quality and safety analysis results, and steps S321-S323 are repeated based on the evolved engineering quality and safety knowledge graph to obtain corresponding evolution analysis results.

[0013] Preferably, in step S32, the evolution is performed based on the current engineering quality and safety knowledge graph and the engineering quality and safety analysis results, including: When the current process node is delayed, an evolutionary analysis is performed on the material nodes associated with the current process node and the next process node whose dependency strength exceeds the standard. The maximum waiting time under the condition that the corresponding material node meets the material performance strength requirements is obtained by evolution, thereby obtaining the required completion time for the current process and the required completion time for the next process node as the evolutionary analysis results; According to the planned completion time of the current process node, the associated material node is evolved to obtain the evolved material performance attenuation value of the material node when the plan is completed, and the evolved material performance attenuation value is further compared with the standard to obtain the evolved material performance analysis result; When the safety node is activated and the safety triggering intensity exceeds the preset standard, the current process will be delayed and evolved to the situation when the current process is delayed.

[0014] Preferably, the abnormal engineering quality analysis results are input into the knowledge graph enhanced large language model, which triggers further evolutionary analysis based on the abnormal engineering quality analysis results, and further obtains the evolved engineering quality analysis results based on the evolved data, thereby forming a logical chain of analysis results; based on the engineering quality analysis results associated with the logical chain, the corresponding judgment basis and processing suggestions are further retrieved from the knowledge base to generate a visual analysis report.

[0015] The beneficial effects of the present invention are: the present invention proposes a digital management method for engineering quality and safety, in which a targeted engineering quality and safety knowledge graph is built based on historical engineering feature data, and in the actual engineering quality management process, the engineering quality and safety knowledge graph is dynamically updated based on real-time engineering site data obtained from the engineering project, and finally the intelligent engineering quality and safety analysis and evolution are realized by enhancing the big language model through the knowledge graph, thereby outputting a visualized engineering quality and safety evaluation result, which helps to improve the pertinence and application effect of the big language model for quality and safety control of engineering projects, and improve the intelligence level of digital management of engineering projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention is further described with reference to the accompanying drawings. However, the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the steps of a digital management method for engineering quality and safety that incorporates a large language model, as shown in an embodiment of the present invention; Figure 2 for Figure 1 Schematic diagram of the specific steps of step S3; Figure 3 for Figure 2 Specific step diagram of step S32 in FIG. DETAILED DESCRIPTION

[0018] The present invention is further described in conjunction with the following application scenarios.

[0019] See also Figure 1 The present invention provides a digital management method for engineering quality and safety that incorporates a large language model, comprising the following steps: S1 extracts engineering feature data from the basic database and builds an engineering quality and safety knowledge graph based on the engineering feature data, where the engineering feature data includes process feature data, material feature data, and safety monitoring data; S2 obtains the engineering site data of the current engineering project; Based on the established engineering quality and safety knowledge graph, S3 conducts engineering quality and safety analysis on engineering site data using the knowledge graph-enhanced large language model and generates engineering quality and safety evaluation results.

[0020] The above-mentioned embodiment of the present invention proposes a digital management method for engineering quality and safety, in which a targeted engineering quality and safety knowledge graph is built based on historical engineering feature data, and in the actual engineering quality management process, the engineering quality and safety knowledge graph is dynamically updated based on real-time engineering site data obtained from the engineering project. Finally, the knowledge graph is enhanced to realize intelligent engineering quality and safety analysis and evolution, thereby outputting a visualized engineering quality and safety evaluation result, which helps to improve the pertinence and application effect of the large language model in quality and safety control of engineering projects, and improve the intelligence level of digital management of engineering projects.

[0021] Among them, the digital management method for engineering quality and safety proposed in the above embodiment can be implemented based on locally deployed smart terminals or cloud servers, so as to provide targeted and visual engineering quality and safety assessment results according to the needs of actual scenarios.

[0022] Preferably, step S1 includes: S11 obtains construction data of historical projects from a basic database, and extracts process characteristic data based on the construction data, wherein the process characteristic data includes process name, construction period information, and process scheduling information; S12 further extracts material characteristic data from the construction data, wherein the material characteristic data includes material type, material performance record data, corresponding process and usage location, etc.; S13 extracts safety monitoring data based on the engineering records of historical projects, where the safety monitoring data includes recorded quality safety events, processes corresponding to the safety events, and rules and clauses associated with the safety events; S14 builds an engineering quality and safety knowledge graph based on the acquired process characteristic data, material characteristic data and safety monitoring data, and completes the setting of experience parameters in the engineering quality and safety knowledge graph.

[0023] The basic database records the construction data of historical projects, including construction plan data, real log data, and monitoring data. By mining the required construction data, material characteristic data, and safety monitoring data from the data recorded in historical projects, it can help to cover the engineering quality and safety knowledge graph and improve the data richness of the knowledge graph. The data in the basic database can be expressed as BIM model data, on-site management logs, or data collected based on on-site monitoring equipment. By storing the relevant data of historical projects in the basic database and building a comprehensive database, the level of data management can be improved, and it is also convenient for subsequent data retrieval and analysis. The basic database also further sets up an enterprise-specific knowledge base required for construction project quality and safety management, and integrates reference knowledge such as laws and regulations related to construction project quality and safety management, engineering construction specifications, procedures, standards and industry-related standards, construction project construction process and specifications and standards, quality and safety construction plans and operating standards, emergency plans, potential quality defects and safety risk identification and control measures, as well as enterprise engineering quality and safety management processes and systems, job responsibilities and work standards, project management standards, quality and safety standardization management manuals, standardized management forms, quality defects and control lists, FMEA analysis databases, hazard sources and dangerous engineering control lists, quality and safety management evaluation indicators and assessments, as well as application cases of BIM technology in digital management of engineering quality and safety, reference knowledge of application cases of PDCA / AHP / FMEA+intelligent algorithm comprehensive models in digital management of engineering quality and safety, etc. into the basic database to facilitate the subsequent provision of large language model calls.

[0024] Among them, the engineering quality and safety knowledge graph proposed in the present invention specifically extracts the required construction data, material characteristic data and safety monitoring data from the basic database, and uses this as the basis to build a three-level knowledge graph based on process-material-safety events, which will help to improve the effect of further safety analysis and evolution based on the knowledge graph, and improve the targeted level of engineering quality and safety management.

[0025] Preferably, the engineering quality and safety knowledge graph constructed includes three types of nodes: process nodes, material nodes, and safety nodes; each process node corresponds to a process, each material node corresponds to a material, and each safety node corresponds to a type of quality and safety event; During the construction phase: Based on the process characteristic data of historical projects, obtain the construction period information of each process node, and calculate the delay probability of each process node based on the construction period time information ,in represents the delay probability of process i, Represents the statistical number of process i in the historical project, represents the number of delays of process i in the historical project; According to the material characteristic data of historical projects, the performance attenuation function corresponding to each material node is calculated ,in represents the performance decay function of material k at time t, represents the initial performance value of material k, represents the decay rate parameter, Indicates the ambient temperature, Indicates the ambient humidity, Represents the protective measure parameters, represents the environmental sensitivity parameter of material k; Based on the safety monitoring data of historical projects, the probability of quality safety incidents corresponding to each safety node causing abnormalities in the process or materials is calculated ,in Indicates a security incident The probability of causing an exception, Indicates security events in historical projects The total number of times recorded, Indicates security events in historical projects The number of times an anomaly occurs in the associated process or material while being recorded.

[0026] Complete the setting of parameters of corresponding nodes based on relevant data of historical projects as the initial engineering quality and safety knowledge graph.

[0027] At the same time, based on the recorded historical safety incident data and process abnormality record information, the probability of the process being triggered by a safety incident is statistically analyzed. ,in Indicates that process i is affected by a security event The probability of triggering Indicates the total number of abnormalities recorded in process i, Indicates the process When an abnormality occurs in the record, a security incident occurs at the same time The total number of times will As the association weight between process node i and safety node v.

[0028] According to the recorded historical security incident record data and material abnormality record information, the probability of the material being triggered by a security incident is calculated ,in Indicates that material k was a security incident The probability of causing an exception, Indicates the total number of abnormalities in material k records. Indicates that a security incident occurs when an abnormality occurs in the material k record The total number of times will As material node k and security node The weight of the association between .

[0029] Based on the data recorded in historical projects, we can first build the node architecture of the engineering quality and safety knowledge graph. The nodes of the engineering quality and safety knowledge graph are divided into three types: process nodes, material nodes, and safety nodes. The process node corresponds to a process in the construction process of the engineering project (such as concrete pouring, formwork installation, foundation treatment, waterproofing, site leveling, etc., and the process information also includes the construction object). The entire process chain can be represented through the one-way path between process nodes. The process node records the knowledge data and experience data related to the corresponding process. The knowledge data includes the construction specifications, cases, and other knowledge data in the knowledge base that match the process (for example, different specifications and standards for the same process can be represented as different process nodes). The experience data includes the proposed delay probability. The process delay probability obtained through experience data statistics can reflect the delay situation estimate of the process based on statistical experience, providing a reference for subsequent further analysis of process abnormality chain analysis and quality and safety management analysis of overall engineering quality control. The material node corresponds to the materials required in the engineering project (such as concrete, steel, wood, water pipes, paint, cement, etc.). Each material node records the knowledge data and experience data related to the corresponding material, among which the knowledge data includes the material standards, instructions for use, etc.; the experience data includes the proposed performance decay function. Taking into account that the performance of the material will change with the stacking time and environmental factors, the data is recorded according to the changes in the material performance data recorded in historical projects, combined with the environmental data at that time, etc., so as to obtain the performance decay function corresponding to the material through fitting, simulation, etc. to reflect the performance changes of the material over time. This can be used as a basis to provide a reference for whether the material performance will cause abnormalities when further combining process delays or engineering quality evolution. A safety node corresponds to a safety event (such as impact, corrosion, high temperature, rainfall, snowfall, etc., which may affect the quality of the project. These events are extracted based on the safety events recorded in the historical records, and events of different degrees can be divided into different safety nodes for further refinement). The safety node records the cases and standards of safety events, as well as the probability that the proposed safety event will trigger abnormalities in other processes or materials. The acquisition of the triggering probability obtained through historical data statistics can provide a basis for subsequent further assessment of whether there will be chain reactions in processes, materials or global impacts when a safety event occurs.

[0030] Based on the engineering quality and safety knowledge graph proposed in the present invention, by extracting and counting historical data, the empirical parameters of relevant nodes in the knowledge graph are improved, which can help to truly reflect the level of mutual influence among process, material and safety events during the construction of engineering projects, so as to adapt to subsequent quality and safety analysis and further evolution based on abnormal situations, thereby providing a basis for possible chain reactions, thereby improving the pertinence and effectiveness of engineering quality and safety assessments.

[0031] The acquisition of specific data can be achieved by performing text parsing, key feature data extraction, and structured processing based on the recorded data, which is not specifically limited in the present invention. For example, for structured data, corresponding historical data can be obtained through completion, interpolation, feature extraction, etc. For unstructured data, such as log text, corresponding feature data can be obtained through text parsing and text extraction.

[0032] Among them, according to the material data of historical projects, the performance attenuation function is corrected by mathematical statistics based on the material data recorded in the historical projects, such as the performance values ​​of the materials at different times, and the corresponding real data such as ambient temperature, ambient humidity, and protection measures level. The classification or fitting method is used to empirically extract the values ​​of the key parameters in the function, such as the attenuation rate parameters, protection measures parameters, and environmental sensitivity parameters, so as to obtain parameter settings that adapt to different situations.

[0033] Preferably, the engineering site data of the current engineering project includes construction plan information, construction log, on-site monitoring data and safety monitoring data; wherein step S2 includes: S21 acquires corresponding planned process characteristic data based on the construction plan information of the current project, wherein the planned process characteristic data includes process name, planned construction period information, planned process scheduling information, and planned process material information, and further acquires actual process characteristic data based on the construction log of the current project, wherein the actual process characteristic data includes process name, actual construction period information, actual process scheduling information, and actual process material consumption information, etc.; S22 obtains material performance parameters and on-site environmental data based on on-site monitoring data of the current project, wherein the material performance parameters are obtained through material test reports, and the on-site environmental data are obtained through sensors installed at the construction site; S23 obtains recorded quality and safety event information based on the safety monitoring data of the current engineering project.

[0034] Based on the established engineering quality and safety knowledge graph, real-time engineering site data of the target engineering project is further collected, and the real real-time engineering site data is further integrated into the engineering quality and safety knowledge graph, so as to obtain a complete engineering quality and safety knowledge graph data for the current situation.

[0035] Preferably, in step S2, the acquired construction site data is aligned to the same timestamp standard, wherein the construction plan information can be extracted for the plan schedule of the entire project and the construction plan schedule of the specific process. By recording information in real-time construction logs, the completion status of the actual process and the corresponding completion time can be updated in real time. At the same time, the material data required for each process is also monitored in real time, and the environmental data of the construction site is also monitored in real time, and these data are updated in real time as the on-site monitoring data. When a safety incident is detected at the construction site, it is updated in real time as the safety monitoring data.

[0036] The material performance parameters can be obtained by measuring using dedicated equipment or methods, and the present invention does not specifically limit this.

[0037] Preferably, see Figure 2 , step S3 includes: S31 updates the engineering site data to the engineering quality and safety knowledge graph based on the acquired engineering site data; S32 performs engineering quality and safety analysis based on the updated safety knowledge graph and the knowledge graph-enhanced large language model to obtain the engineering quality and safety analysis results; S33 performs transformation based on the obtained engineering quality and safety analysis results to obtain a visualized engineering quality and safety evaluation result.

[0038] When conducting further quality and safety analysis based on engineering site data, the engineering prior data is first updated according to the classification of processes, materials and safety events to the information of the corresponding process nodes, material nodes and safety nodes in the engineering quality and safety knowledge graph, so as to obtain a real-time and dynamic engineering quality and safety knowledge graph. Based on the real-time engineering quality and safety knowledge graph, the engineering quality and safety analysis is performed through the enhanced language big model (such as real-time analysis, evolutionary analysis, local analysis, overall analysis, etc.), so as to obtain the corresponding engineering quality and safety analysis results and further convert them into text output to obtain visual engineering quality and safety evaluation results. Based on the above method, the real-time and targeted engineering quality and safety analysis and evaluation of the current engineering project are completed, which can improve the intelligence level and effect of the digital management of engineering quality and safety of the engineering project.

[0039] Preferably, in step S31, updating the project site data to the project quality and safety knowledge graph specifically includes: In the initialization phase: update the corresponding process nodes according to the obtained planned process feature data, activate and connect the corresponding process nodes to form a process chain, where each process node contains the planned start time and planned end time of the process, the related process node information and the required material information; adjust the association weight between the process node and the corresponding material node according to the required material information ,in Indicates process node With material nodes The association weight between ; In the dynamic update stage: update the process nodes according to the actual process feature data obtained, update the process status and actual duration information to the corresponding process nodes; further obtain the dependency strength between the associated process nodes in the process chain ,in Indicates the dependence strength of process node j on process node i; Represents the delay probability of process node i. When process node i is delayed compared with the planned end time, ,otherwise , It represents the minimum waiting time of process node j after process node i is completed; represents the time length between the planned / actual completion time of process node i and the planned start time of process node j; Activate and update the material nodes based on the obtained on-site monitoring data, update the material performance parameters and on-site environmental data to the material nodes, and further estimate the real-time material performance attenuation value ,in Represents the material performance attenuation value of material k after time t, where t represents the time difference between the current moment and the initial performance value recorded by the material, Indicates the initial performance value of material k recorded, obtained according to the test report, Indicates the ambient temperature, Indicates the ambient humidity, which is obtained based on the data collected by the environmental sensor. Represents the protective measure parameter, obtained according to the current protective measures for material k, represents the environmental sensitivity parameter of material k; Update the safety node based on the obtained safety monitoring data, activate the corresponding safety node based on the recorded quality safety event, and record the quality safety event information in the safety node, where the quality safety event information includes the type of safety event, the time of occurrence, the corresponding material or process information, and the associated regulatory provisions; further calculate the safety triggering strength of the safety node ,in Indicates a security incident The probability of causing an exception, Indicates the process currently in progress Security incident The probability of triggering Indicates activation parameters. When security events are recorded, otherwise .

[0040] Among them, when the association weight between two nodes is greater than 0, it means that there is an association between the two nodes.

[0041] The updating of the engineering quality and safety knowledge graph includes two stages. One is that after the construction plan of the project is completed, the corresponding process nodes in the engineering quality and safety knowledge graph can be activated according to the construction plan information. At the same time, according to the specific construction information, the material nodes corresponding to the process nodes are further activated to obtain an initial stage process chain. The other stage is after the actual construction begins, the corresponding nodes are updated according to the engineering site data obtained in real time, so that the corresponding nodes contain the latest data and information. The real-time update of the process nodes is to update the completion status of the process nodes at the current time to the process nodes, so as to determine whether the process nodes are delayed or the specific delay time. The dependency strength of the process nodes is updated according to the specific time information. The higher the dependency strength, the higher the dependency between the related process nodes. In this way, once the previous process node is delayed, it is likely to cause a chain of abnormal situations. For material nodes, the current performance parameters of the material nodes are updated based on the latest monitored material performance parameters (such as concrete strength, concrete stress, steel corrosion resistance, cement strength, etc.), and are also updated based on the acquired environmental information. This allows the material nodes to feedback the current status of the corresponding materials, providing a real basis for further analysis and evolution. For safety nodes, when a corresponding safety incident is detected based on the project site data, the corresponding safety node is activated and the relevant information of the safety incident is recorded in the corresponding safety node. According to the above method, the engineering quality and safety knowledge map can be updated in real time and dynamically, improving the pertinence and reliability of subsequent engineering quality and safety analysis and evaluation.

[0042] Preferably, see Figure 3 In step S32, engineering quality and safety analysis is performed based on the knowledge graph-enhanced large language model, including: S321 detects the dependency strength between the current process nodes, compares the dependency strength with the set dependency standard value, and obtains the delay risk analysis result; S322 detects the material performance attenuation value of the current material node, compares the material performance attenuation value with the set material performance standard value, and obtains the material performance analysis result; S323 detects the security triggering strength of the current security node, compares the security triggering strength with the set security standard value, and obtains the security event analysis result.

[0043] In one scenario, the knowledge graph-enhanced large language model can call the built-in engineering specification database and match the corresponding standards based on the node information that needs to be analyzed to complete the corresponding comparison operations and obtain the corresponding analysis results.

[0044] In one scenario, the process dependency strength of the current process node is obtained through process node analysis. , indicating that the current dependency between the two processes is too high, and it is necessary to focus on monitoring other nodes associated with the two processes.

[0045] In one scenario, the current concrete strength is obtained through material node information If it is less than the set threshold of 0.7, the analysis result of insufficient concrete strength is obtained.

[0046] In one scenario, when the security trigger strength is obtained based on the security node information When it exceeds the preset threshold of 0.1, the material nodes associated with the current process node or safety node will be further monitored and further evolution analysis will be activated.

[0047] Based on the deduction capability of the large language model, further deduction is performed based on the current engineering quality and safety knowledge graph to obtain further evolutionary analysis results.

[0048] In one scenario, the deduction method includes evolving the delay of the currently ongoing process, evolving the maximum waiting time of the materials associated with the process, evolving whether the maximum waiting time of the materials in the case of process delays meets the completion time requirements of the corresponding process, etc.

[0049] Preferably, step S32 further includes: S324 evolves based on the current engineering quality and safety knowledge graph and engineering quality and safety analysis results, and repeats steps S321-S323 based on the evolved engineering quality and safety knowledge graph to obtain corresponding evolutionary analysis results.

[0050] Preferably, in step S32, the evolution is performed based on the current engineering quality and safety knowledge graph and the engineering quality and safety analysis results, including: When the current process node is delayed, an evolutionary analysis is performed on the material nodes associated with the current process node and the next process node whose dependency strength exceeds the standard. The maximum waiting time under the condition that the corresponding material node meets the material performance strength requirements is obtained by evolution, thereby obtaining the required completion time for the current process and the required completion time for the next process node as the evolutionary analysis results; According to the planned completion time of the current process node, the associated material node is evolved to obtain the evolved material performance attenuation value of the material node when the plan is completed, and the evolved material performance attenuation value is further compared with the standard to obtain the evolved material performance analysis result; When the safety node is activated and the safety triggering intensity exceeds the preset standard, the current process will be delayed and evolved to the situation when the current process is delayed.

[0051] The above-mentioned evolutionary analysis method can also be used to further perform evolutionary analysis based on the results obtained by the evolutionary analysis to obtain comprehensive analysis results.

[0052] In one scenario, depending on the actual situation, the evolution analysis also includes: The material nodes are evolved to obtain the maximum waiting time when the corresponding material nodes meet the material performance strength requirements, so as to determine whether the maximum waiting time meets the planned construction time of the associated process.

[0053] By using a large language model to perform evolutionary analysis on the current engineering quality and safety knowledge graph, a logical chain can be formed to accurately simulate the evolution of possible chain reactions, thereby obtaining estimated evolutionary analysis results. Compared with traditional engineering quality and safety analysis methods, this logical chain-based analysis method can improve the adaptability and accuracy of evolutionary analysis, helping to enhance the logical and intelligent level of engineering quality and safety analysis and assessment. Furthermore, the evolutionary analysis results obtained are based on real engineering projects, making them highly adaptable and targeted, further improving the effectiveness of engineering quality and safety analysis.

[0054] At the same time, the evolution method based on the logical chain can improve the generation effect of early warning and response plans for abnormal situations, and improve the effect of digital management of engineering quality and safety based on large language models.

[0055] Preferably, step S32 includes: Conduct comprehensive risk analysis based on the process nodes included in the process chain and the material nodes and safety nodes associated with the process nodes to calculate the total risk value of the engineering project ,in Represents the estimated total risk value of the engineering project, Represented as consecutive process nodes in the process chain, represents the dependence strength of process node j on process node i, Represented as the associated material node k in the process chain, Indicates the material performance attenuation value of material k at the current moment, represents the initial property value recorded for material k, Represented as an associated safety node in the process chain , Represents a secure node The safety triggering strength, 、 、 Respectively represent the set weight factors; The comprehensive risk analysis results are obtained by comparing the total risk value of the engineering project with the set risk standards.

[0056] At the same time, based on the evolutionary or real-time engineering quality and safety knowledge graph, we can further conduct a comprehensive risk analysis of the overall engineering project and provide a basis for engineering quality and safety management.

[0057] In one scenario, the specific steps of performing engineering quality safety analysis based on the updated safety knowledge graph in step S32 can be implemented based on the internal reasoning module of the large language model, and the corresponding steps can be implemented by enhancing the large language model with the knowledge graph, or by setting up and training a special deep learning model. The present invention does not make specific limitations here.

[0058] Preferably, step S33 includes: Based on the obtained analysis results, the engineering quality and safety analysis results and the corresponding node information, input parameters are generated, and the input parameters are input into the knowledge graph enhanced large language model to obtain a visualized engineering quality and safety evaluation result; wherein the engineering quality and safety analysis result includes at least one of the delay risk analysis result, material performance analysis result, safety incident analysis result and comprehensive risk analysis result.

[0059] Preferably, step S33 includes: The abnormal engineering quality analysis results are input into the knowledge graph enhanced large language model, which triggers further evolutionary analysis based on the abnormal engineering quality analysis results, and further obtains the evolved engineering quality analysis results based on the evolved data, thereby forming a logical chain of analysis results; according to the engineering quality analysis results associated with the logical chain, the corresponding judgment basis and processing suggestions are further retrieved from the knowledge base to generate a visual analysis report.

[0060] Based on the large language model, a visual engineering quality and safety assessment report is further generated. According to the engineering quality and safety analysis results, the corresponding knowledge can be retrieved from the standard database to further explain and display the analysis results. This helps managers propose engineering quality and safety improvement plans based on the assessment report and realize digital management of engineering quality and safety.

[0061] In an exemplary scenario, based on the current engineering quality and safety knowledge graph, when a safety event is detected (such as a rainstorm with a 24-hour rainfall of 60mm that has lasted for 15 minutes), the safety triggering strength obtained based on the safety node exceeds the standard value. It is judged that the probability of the current process (such as concrete pouring) being delayed due to the abnormality increases. That is, according to the current engineering quality and safety knowledge graph, the delay of the current process is evolved, and the related material nodes are further evolved. The maximum waiting time of the current material node (concrete) is evolved to 30 minutes; at the same time, the dependency strength of the current process node is further analyzed. If it is less than the set standard of 0.6, it indicates that the next process node (maintenance) has a low degree of dependence on the current process node. In the event of a delay in the current process node, the next process node will not be greatly affected. According to the above evolution results, the evolution analysis results corresponding to the obtained logic chain are: heavy rain occurs → concrete pouring is affected → concrete pouring is delayed → the maximum waiting time for concrete is 30 minutes. The resulting treatment suggestions are: 1. Take rain prevention measures; 2. Improve concrete quality monitoring requirements; 3. Complete pouring within 30 minutes; 4. Reserve concrete materials.

[0062] In another exemplary scenario, based on the current engineering quality and safety knowledge graph, when a safety incident is detected (such as insufficient concrete strength of the structure in area B), the safety triggering strength obtained based on the safety node exceeds the standard value, and the probability of the current process (such as concrete pouring in area B) being delayed due to the abnormality is judged to be increased. That is, according to the current engineering quality and safety knowledge graph, the situation of the current process being delayed (delayed for 30 days) is evolved, and the dependence strength of the current process node on the subsequent process (such as concrete pouring in area D) is further analyzed. ,…, If both are greater than the set standard of 0.6, it indicates that the subsequent process nodes are all affected by the delay. For process node q, its associated material node (cement) is further obtained, and the material node is evolved, and the maximum waiting time for the material node is 20 days. According to the above evolution results, the evolution analysis results corresponding to the obtained logic chain are: insufficient concrete strength of area B → 30-day delay in concrete pouring in area B → overall construction period delay → maximum waiting time of cement for concrete pouring in area D is 20 days... The resulting treatment suggestions are: 1. Stop work immediately (according to the "Concrete Structure Engineering Construction Quality Acceptance Code": "Insufficient concrete strength requires immediate suspension of work for inspection"), and complete the reinforcement of the concrete in area B (with corresponding accident cases and reinforcement cases); 2. Adjust the subsequent process plan (with intelligent adjustment results); 3. Complete the concrete pouring in area D within 20 days; 4. Prepare spare cement materials.

[0063] From the above description of the embodiments, those skilled in the art will appreciate that the embodiments described herein can be implemented using hardware, software, firmware, middleware, code, or any appropriate combination thereof. For hardware implementation, the processor may be implemented in one or more of the following: an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field-programmable gate array (FPGA), a processor, a controller, a microcontroller, a microprocessor, or other electronic units designed to implement the functionality described herein, or any combination thereof. For software implementation, some or all of the processes of the embodiments may be performed by a computer program instructing the relevant hardware. During implementation, the program may be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media includes any medium that facilitates the transfer of a computer program from one location to another. A storage medium may be any available medium that can be accessed by a computer. Computer-readable media may include, but are not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should analyze that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A digital management method for engineering quality and safety that incorporates a large language model, characterized in that: The steps include: S1 extracts engineering feature data from the basic database and builds an engineering quality and safety knowledge graph based on the engineering feature data, where the engineering feature data includes process feature data, material feature data, and safety monitoring data; S2 obtains the engineering site data of the current engineering project; Based on the established engineering quality and safety knowledge graph, S3 conducts engineering quality and safety analysis on engineering site data using the knowledge graph-enhanced large language model and generates engineering quality and safety evaluation results.

2. The digital management method for engineering quality and safety integrating a large language model according to claim 1 is characterized in that: Step S1 includes: S11 obtains construction data of historical projects from a basic database, and extracts process characteristic data based on the construction data, wherein the process characteristic data includes process name, construction period information, and process scheduling information; S12 further extracts material characteristic data from the construction data, wherein the material characteristic data includes material type, material performance record data, corresponding process and usage location; S13 extracts safety monitoring data based on engineering records of historical projects, where the safety monitoring data includes recorded quality safety events, processes corresponding to the safety events, and rules and clauses associated with the safety events; S14 builds an engineering quality and safety knowledge graph based on the acquired process characteristic data, material characteristic data and safety monitoring data, and completes the setting of experience parameters in the engineering quality and safety knowledge graph.

3. The digital management method for engineering quality and safety integrating a large language model according to claim 2 is characterized in that: The engineering quality and safety knowledge graph we built includes three types of nodes: process nodes, material nodes, and safety nodes. Each process node corresponds to a process, each material node corresponds to a material, and each safety node corresponds to a type of quality and safety event. During the construction phase: Based on the process characteristic data of historical projects, obtain the construction period information of each process node, and calculate the delay probability of each process node based on the construction period time information ,in represents the delay probability of process i, Represents the statistical number of process i in the historical project, represents the number of delays of process i in the historical project; According to the material characteristic data of historical projects, the performance attenuation function corresponding to each material node is calculated ,in represents the performance decay function of material k at time t, represents the initial performance value of material k, represents the decay rate parameter, Indicates the ambient temperature, Indicates the ambient humidity. Represents the protective measure parameters, represents the environmental sensitivity parameter of material k; Based on the safety monitoring data of historical projects, the probability of quality safety incidents corresponding to each safety node causing abnormalities in the process or materials is calculated ,in Indicates a security incident The probability of causing an exception, Indicates security events in historical projects The total number of times recorded, Indicates security events in historical projects The number of times an anomaly occurs in the associated process or material while being recorded.

4. The digital management method for engineering quality and safety integrating a large language model according to claim 3 is characterized in that: The engineering site data of the current engineering project includes construction plan information, construction logs, on-site monitoring data, and safety monitoring data; wherein step S2 includes: S21 acquires corresponding planned process characteristic data based on the construction plan information of the current project, wherein the planned process characteristic data includes process name, planned construction period information, planned process scheduling information, and planned process required material information. Furthermore, actual process characteristic data is acquired based on the construction log of the current project, wherein the actual process characteristic data includes process name, actual construction period information, actual process scheduling information, and actual process consumed material information. S22 obtains material performance parameters and on-site environmental data based on on-site monitoring data of the current project, wherein the material performance parameters are obtained through material test reports, and the on-site environmental data are obtained through sensors installed at the construction site; S23 obtains recorded quality and safety event information based on the safety monitoring data of the current engineering project.

5. The digital management method for engineering quality and safety integrating a large language model according to claim 4 is characterized in that: Step S3 includes: S31 updates the engineering site data to the engineering quality and safety knowledge graph based on the acquired engineering site data; S32 performs engineering quality and safety analysis based on the updated safety knowledge graph and the knowledge graph-enhanced large language model to obtain the engineering quality and safety analysis results; S33 performs transformation based on the obtained engineering quality and safety analysis results to obtain a visualized engineering quality and safety evaluation result.

6. The method for digital management of engineering quality and safety incorporating a large language model according to claim 5 is characterized in that: In step S31, the project site data is updated to the project quality and safety knowledge graph, specifically including: In the initialization phase: update the corresponding process nodes according to the obtained planned process feature data, activate and connect the corresponding process nodes to form a process chain, where each process node contains the planned start time and planned end time of the process, the related process node information and the required material information; adjust the association weight between the process node and the corresponding material node according to the required material information ,in Indicates process node With material nodes The association weight between ; In the dynamic update stage: update the process nodes according to the actual process feature data obtained, update the process status and actual duration information to the corresponding process nodes; further obtain the dependency strength between the associated process nodes in the process chain ,in Indicates the dependence strength of process node j on process node i; Represents the delay probability of process node i. When process node i is delayed compared with the planned end time, ,otherwise , It represents the minimum waiting time of process node j after process node i is completed; represents the length of time between the planned / actual completion time of process node i and the planned start time of process node j; Activate and update the material nodes based on the obtained on-site monitoring data, update the material performance parameters and on-site environmental data to the material nodes, and further estimate the real-time material performance attenuation value ,in Represents the material performance attenuation value of material k after time t, where t represents the time difference between the current moment and the initial performance value recorded by the material, Indicates the initial performance value of material k recorded, obtained according to the test report, Indicates the ambient temperature, Indicates the ambient humidity, which is obtained based on the data collected by the environmental sensor. Represents the protective measure parameter, obtained according to the current protective measures for material k, represents the environmental sensitivity parameter of material k; Update the safety node based on the obtained safety monitoring data, activate the corresponding safety node based on the recorded quality safety event, and record the quality safety event information in the safety node, where the quality safety event information includes the type of safety event, the time of occurrence, the corresponding material or process information, and the associated regulatory provisions; further calculate the safety triggering strength of the safety node ,in Indicates a security incident The probability of causing an exception, Indicates the process currently in progress Security incident The probability of triggering Indicates activation parameters. When security events are recorded, otherwise .

7. The digital management method for engineering quality and safety integrating a large language model according to claim 6 is characterized in that: In step S32, engineering quality and safety analysis is performed based on the knowledge graph-enhanced large language model, including: S321 detects the dependency strength between the current process nodes, compares the dependency strength with the set dependency standard value, and obtains the delay risk analysis result; S322 detects the material performance attenuation value of the current material node, compares the material performance attenuation value with the set material performance standard value, and obtains the material performance analysis result; S323 detects the security triggering strength of the current security node, compares the security triggering strength with the set security standard value, and obtains the security event analysis result; S324 evolves based on the current engineering quality and safety knowledge graph and engineering quality and safety analysis results, and repeats steps S321-S323 based on the evolved engineering quality and safety knowledge graph to obtain corresponding evolutionary analysis results.

8. The digital management method for engineering quality and safety integrating a large language model according to claim 7 is characterized in that: In step S32, the current engineering quality and safety knowledge graph and engineering quality and safety analysis results are evolved, including: When the current process node is delayed, an evolutionary analysis is performed on the material nodes associated with the current process node and the next process node whose dependency strength exceeds the standard. The maximum waiting time under the condition that the corresponding material node meets the material performance strength requirements is obtained by evolution, thereby obtaining the required completion time for the current process and the required completion time for the next process node as the evolutionary analysis results; According to the planned completion time of the current process node, the associated material node is evolved to obtain the evolved material performance attenuation value of the material node when the plan is completed, and the evolved material performance attenuation value is further compared with the standard to obtain the evolved material performance analysis result; When the safety node is activated and the safety triggering intensity exceeds the preset standard, the current process will be delayed and evolved to the situation when the current process is delayed.

9. The digital management method for engineering quality and safety integrating a large language model according to claim 8 is characterized in that: The abnormal engineering quality analysis results are input into the knowledge graph enhanced large language model, which triggers further evolutionary analysis based on the abnormal engineering quality analysis results, and further obtains the evolved engineering quality analysis results based on the evolved data, thereby forming a logical chain of analysis results; according to the engineering quality analysis results associated with the logical chain, the corresponding judgment basis and processing suggestions are further retrieved from the knowledge base to generate a visual analysis report.

Images