Digitalized construction management system

Abstract

This invention discloses a digitally-based construction management system, belonging to the field of digital management technology. It includes: constructing a network of low-voltage electrical construction process nodes according to subway engineering design specifications; collecting real-time actual construction data and timestamps of each node and verifying their continuity and consistency, generating process deviation early warning signals; marking abnormal nodes based on the detection results and performing bidirectional tracing to locate the source process of the abnormality and subsequent affected processes; dynamically optimizing the process deviation early warning threshold and generating an adaptive monitoring strategy based on the process deviation early warning signals and abnormal tracing results; generating a construction management report containing abnormal process nodes, abnormality types, tracing paths, and associated early warnings, and solidifying it through digital signatures and trusted timestamps. This invention achieves transparent monitoring of the construction process, rapid abnormal tracing, and intelligent early warning, improving the efficiency and traceability of construction quality management.

Description

Technical Field

[0001] This invention relates to the field of digital management technology, and more specifically to a digitally based construction management system. Background Technology

[0002] In the field of urban rail transit construction, especially in the construction of subway low-voltage electrical systems, the construction quality of key processes such as cable laying, equipment installation, and joint fabrication directly affects the reliability and safety of the overall system. Because the construction of low-voltage electrical systems involves numerous procedures and complex dependencies, any anomaly such as exceeding construction parameters or timing misalignment in any stage can directly impact subsequent processes, potentially leading to system performance degradation or even functional failure. Therefore, meticulous and real-time monitoring and management of the entire construction process, along with rapid location and impact assessment of anomalies, is a core requirement for ensuring the quality of subway low-voltage electrical system projects.

[0003] Currently, the management of such construction processes mainly relies on paper or discrete electronic records kept by construction units, and irregular on-site inspections by supervisors. This method has significant shortcomings: First, information from each process is recorded in a scattered manner, forming information silos that are difficult to integrate and compare in real time. Second, when an anomaly occurs in a certain stage, the lack of a digital mapping of inter-process dependencies and chain-like data association makes it difficult to quickly and accurately trace back to the responsible upstream process, and also makes it impossible to effectively assess the potential impact of the anomaly on downstream processes, leading to a break in the traceability chain. This not only results in low efficiency in problem localization and prolongs the troubleshooting and repair cycle, but also easily leads to disputes afterward due to unclear liability definitions. Summary of the Invention

[0004] This invention addresses the technical problems in existing technologies, such as information silos in the construction process, difficulty in quickly tracing and assessing anomalies, unclear definition of responsibilities, and low management efficiency, by providing a digitally based construction management system.

[0005] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0006] In a first aspect, the present invention provides a digitally based construction management system, comprising:

[0007] The process network construction module is used to establish a network of low-voltage construction process nodes according to the subway engineering design specifications. Each process node includes the process type, technical parameter thresholds, and inter-process dependencies.

[0008] The construction data verification module is used to collect actual construction data and construction timestamps for each process node, and verify continuity and consistency. When a deviation is detected, a process deviation warning signal is generated.

[0009] The anomaly tracing and location module is used to mark abnormal process nodes based on the detection results, and to perform bidirectional tracing along the network of low-voltage construction process nodes to locate the source process of the anomaly and subsequent affected processes.

[0010] The parameter dynamic optimization module is used to dynamically optimize the process deviation warning threshold based on the process deviation warning signal and the abnormality tracing result, generate an adaptive monitoring strategy, and update it to the weak current construction process node network.

[0011] The report generation and consolidation module is used to generate construction management reports that include abnormal process nodes, abnormal types, traceability paths and related early warnings, and to solidify them with digital signatures and trusted timestamps.

[0012] Secondly, the present invention provides a digital-based construction management method, including:

[0013] According to the subway engineering design specifications, a network of low-voltage electrical construction process nodes is established, in which each process node includes process type, technical parameter thresholds and inter-process dependencies.

[0014] Collect actual construction data and construction timestamps for each process node, and verify continuity and consistency. When a deviation is detected, generate a process deviation warning signal.

[0015] Based on the test results, abnormal process nodes are marked, and bidirectional tracing is performed along the network of low-voltage construction process nodes to locate the source process of the abnormality and subsequent affected processes.

[0016] Based on the process deviation early warning signal and the anomaly tracing results, the process deviation early warning threshold is dynamically optimized, an adaptive monitoring strategy is generated, and updated to the weak current construction process node network.

[0017] Generate a construction management report that includes abnormal process nodes, abnormal types, traceability paths, and associated warnings, and solidify it with digital signatures and trusted timestamps.

[0018] The beneficial effects of this invention are:

[0019] Compared to existing technologies, this invention firstly achieves structured integration and visualization of construction information by digitally modeling the entire low-voltage electrical construction process and constructing a network of process nodes, breaking down information silos. Secondly, through real-time data verification and intelligent early warning, it can promptly detect timing deviations and parameter anomalies during construction. Thirdly, based on the constructed network, bidirectional traceability can quickly locate the source of anomalies and assess their potential impact on downstream processes, solving the problem of broken traceability chains in traditional methods and improving the efficiency of problem investigation and responsibility determination. Finally, the system has adaptive optimization and report solidification capabilities, which can continuously improve monitoring accuracy and generate authoritative management evidence, thereby promoting the overall transformation and upgrading of construction management towards intelligence and refinement, ensuring project quality and construction efficiency. Attached Figure Description

[0020] Figure 1 A schematic diagram of the structure of the digital-based construction management system provided by the present invention;

[0021] Figure 2 This is a schematic diagram illustrating the principle of the digital-based construction management system provided by the present invention.

[0022] In the attached diagram, the components represented by each number are as follows:

[0023] The module includes: process network construction module 11, construction data verification module 12, anomaly tracing and location module 13, parameter dynamic optimization module 14, and report generation and solidification module 15. Detailed Implementation

[0024] Example 1, as Figure 1 , Figure 2 As shown, this embodiment of the invention provides a digitally based construction management system, including:

[0025] The process network construction module 11 is used to establish a network of low-voltage construction process nodes according to the subway engineering design specifications. Each process node includes process type, technical parameter threshold and inter-process dependency relationship.

[0026] First, a network of low-voltage electrical system construction process nodes is established based on the subway engineering design specifications. The subway engineering design specifications are a system of standard documents guiding subway engineering construction, including overall engineering design requirements, functional and performance indicators of various professional systems, construction process standards, and quality acceptance criteria. Establishing a network of low-voltage electrical system construction process nodes based on these specifications is a digital and structured mapping of the entire low-voltage system construction process. It transforms complex physical construction procedures, technical requirements, and logical relationships into a set of nodes and links that can be recognized and processed by computers.

[0027] Specifically, by establishing a unified digital model, construction information that was originally scattered across various drawings, documents, and records can be effectively integrated, solving the problem of information fragmentation in traditional management models. This provides a clear, logically coherent, and standardized processing framework for subsequent stages, including real-time data collection, multi-dimensional verification, anomaly tracing, and in-depth analysis.

[0028] Specifically, based on subway engineering design specifications, a network of low-voltage electrical construction process nodes is established, including:

[0029] Extract the cable laying, equipment installation, and connector fabrication procedures defined in the subway low-voltage construction drawings and process documents, and determine them as the procedure types in the procedure node set;

[0030] According to the construction acceptance standards, corresponding technical parameter thresholds are set for each process type, including physical parameter thresholds, signal parameter thresholds, and functional parameter thresholds.

[0031] Analyze the construction process sequence between each process, establish the temporal and logical dependencies between process nodes, and form a network topology with directionality and hierarchy;

[0032] Based on the process types, the technical parameter thresholds corresponding to each process type, and the dependencies between process nodes, a network of process nodes for low-voltage electrical construction is constructed.

[0033] First, the key process definitions explicitly stated in the subway low-voltage electrical construction drawings and process documents are extracted. These drawings and documents serve as the direct basis for project implementation, detailing specific work content such as cable laying, equipment installation, and connector fabrication. These are identified and defined as several process types constituting a set of process nodes, thus transforming the construction requirements in paper or electronic documents into structured data units that can be identified and processed in subsequent processes.

[0034] Secondly, based on the construction acceptance standards, corresponding technical parameter thresholds are configured for each of the aforementioned process types. Specifically, technical parameter thresholds are quantitative indicators for measuring whether construction quality meets standards, and their settings are derived from regulatory requirements and engineering practice experience. Preferably, this technical parameter threshold system covers three dimensions: physical parameter thresholds, signal parameter thresholds, and functional parameter thresholds. For example, for the cable laying process, the physical parameter threshold can specify that the minimum bending radius of the coaxial cable is not less than 10 times the cable's outer diameter; the signal parameter threshold can specify that the attenuation at the single-mode fiber optic splice point is not greater than 0.1 dB; and the functional parameter threshold can require that the cable identification accuracy reaches 100%. For the camera installation and wiring process, the physical parameter threshold can limit the vertical deviation of the mounting bracket to no more than 2°; the signal parameter threshold can require that the output power of the Ethernet power supply port is not less than 12.95 watts; and the functional parameter threshold can specify that the video stream encoding format is H.265. By presetting clear and quantifiable technical parameter thresholds for each process type, precise standards can be provided for the compliance verification of subsequent construction data.

[0035] Furthermore, this analysis examines the sequential connections and logical constraints of each specific construction process within the actual construction workflow, i.e., the construction process sequence. Based on the construction process flow, technical briefing documents, and engineering experience, this analysis aims to clarify two core dependencies between process nodes: temporal dependencies and logical dependencies. Specifically, temporal dependencies refer to the sequential order of processes over time; for example, cable tray installation must be completed before cable laying can proceed. Logical dependencies, on the other hand, refer to the dependence of processes on technical conditions or states; for example, equipment power-on testing can only be performed after all power and signal cable connections are completed and insulation tests are passed.

[0036] By analyzing and establishing temporal and logical dependencies, discrete process nodes can be connected to form a network topology with clear directionality and internal hierarchy, thereby truly reflecting the dynamic logic of the construction process.

[0037] Finally, based on the process types, technical parameter thresholds corresponding to each process, and dependencies between process nodes determined in the preceding steps, a comprehensive integration is performed to formally construct a complete low-voltage electrical construction process node network. This low-voltage electrical construction process node network, as a unified digital model, fully encapsulates all technical requirements, quality standards, and process logic for low-voltage system construction, providing a core data structure and rule foundation for the entire construction management process.

[0038] The construction data verification module 12 is used to collect the actual construction data and construction timestamps of each process node, and verify the continuity and consistency. When a deviation is detected, a process deviation warning signal is generated.

[0039] Since the low-voltage electrical construction process node network has preset technical parameter thresholds for each process node and clarified the dependencies between processes, the core purpose of this construction data verification module is to dynamically compare the real-time data generated at the construction site with the standards and rules defined in the network model. By collecting and verifying actual construction data and construction timestamps, it is possible to instantly determine whether construction activities strictly follow the preset process flow and quality requirements, thereby achieving proactive monitoring of the continuity and consistency of the construction process.

[0040] Specifically, connecting static digital models with dynamic construction practices allows for the timely detection of potential deviations and the prevention of quality defects from spreading. When actual data or time-series relationships deviate from preset rules, a process deviation warning signal containing specific deviation information will be automatically generated, providing a clear and timely triggering basis for subsequent anomaly location and handling.

[0041] Specifically, actual construction data and timestamps for each process node are collected, and continuity and consistency are verified. When a deviation is detected, a process deviation warning signal is generated, including:

[0042] Collect actual construction data and corresponding construction timestamps for each process node, wherein the actual construction data includes measured parameter values ​​of the same type as the technical parameter threshold;

[0043] Based on the temporal dependencies in the inter-process dependencies, verify whether the construction timestamp of each process node is later than the latest completion timestamp of all preceding dependent process nodes;

[0044] Based on the logical dependencies in the inter-process dependencies and the technical parameter thresholds, verify whether the measured parameter values ​​in the actual construction data of each process node meet the corresponding threshold range;

[0045] When time sequence continuity verification or data logic consistency verification fails, a process deviation warning signal is generated, including:

[0046] When timing continuity verification fails, a timing deviation warning signal is generated, which includes the timing deviation type, time offset, and identifier of the involved process node.

[0047] When data logic consistency verification fails, a parameter deviation warning signal is generated, which includes the parameter deviation type, the deviation of the measured parameter value from the technical parameter threshold, and the identifier of the involved process node.

[0048] The timing deviation warning signal and / or the parameter deviation warning signal are bound and encapsulated with the corresponding construction timestamp and process node information to form the process deviation warning signal.

[0049] First, the actual construction data and its corresponding timestamps for each process node are collected. Actual construction data refers to real-time information acquired at the construction site through sensors, manual input, or automated testing equipment. Its content corresponds to the technical parameter threshold categories preset by the process network construction module for each process type. For example, actual construction data may include specific parameters of equipment installation, the physical status of line connections, and real-time results of functional tests. The construction timestamp precisely records the moment each data point was generated.

[0050] Secondly, temporal continuity verification is performed based on the temporal dependencies defined in the low-voltage electrical construction process node network. This verification checks each process node to ensure that its construction timestamp is strictly later than the latest completion timestamps of all its preceding dependent process nodes. This temporal continuity verification step aims to ensure that the construction process is executed strictly in the preset sequence, preventing logical errors and safety hazards caused by skipping steps or performing reverse construction.

[0051] Secondly, data logical consistency verification is performed based on the logical dependencies defined in the low-voltage electrical construction process node network and the preset technical parameter thresholds. This verification, for each process node, compares the measured parameter values ​​included in its actual construction data with the technical parameter threshold range specified for the corresponding process type. For example, for the cable laying process, it is necessary to verify whether the measured cable bending radius meets the minimum bending radius threshold; for the equipment installation process, it is necessary to verify whether the measured installation verticality is within the allowable deviation range. This data logical consistency verification step aims to ensure that the output quality of each construction activity meets the design specifications and acceptance standards.

[0052] If either the sequence continuity verification or the data logic consistency verification fails, a corresponding process deviation warning signal will be automatically generated.

[0053] Specifically, if the timing continuity verification fails, indicating a disruption in the construction sequence, a timing deviation warning signal is generated. This signal explicitly includes the timing deviation type, time offset, and the identifier of the involved process node. If the data logic consistency verification fails, indicating that the construction parameters exceed the allowable range, a parameter deviation warning signal is generated. This signal explicitly includes the parameter deviation type and the degree of deviation of the measured parameter value from the technical parameter threshold.

[0054] Finally, the generated timing deviation warning signals and / or parameter deviation warning signals are bound and encapsulated with their corresponding construction timestamps and complete process node information to form a complete and clearly defined process deviation warning signal. This process deviation warning signal not only indicates the existence of the deviation but also provides key contextual information such as the nature, degree, location, and time of the deviation, providing a precise data foundation for subsequent anomaly tracing and handling decisions.

[0055] The anomaly tracing and location module 13 is used to mark abnormal process nodes according to the detection results, and to perform bidirectional tracing along the network of weak current construction process nodes to locate the source process of the anomaly and subsequent affected processes.

[0056] In the construction process, each work step forms a tightly linked chain based on pre-defined dependencies. An anomaly at a single work step is often not an isolated event; it may be caused by defects left over from upstream processes. Furthermore, this anomaly can propagate down the process chain to downstream processes, amplifying quality risks. Therefore, simply identifying the anomaly step is insufficient to fully reveal the root cause and potential consequences of the problem, and cannot support effective corrective and preventative measures.

[0057] Therefore, the purpose of this anomaly tracing and localization module is to perform in-depth correlation and causal analysis on anomalies identified by the construction data verification module, based on an established and complete network of low-voltage construction process nodes that includes temporal and logical dependencies. On the one hand, it traces backward along the dependencies until it locates the initial anomaly source process that no longer has any pre-existing dependencies, thereby clarifying the responsibility at its root. On the other hand, it explores forward along the dependencies, identifying all subsequent process nodes that may be affected by the current anomaly, thereby accurately defining the scope of the anomaly's impact.

[0058] The aforementioned two-way tracing mechanism enables a comprehensive analysis from a single anomaly point to the complete anomaly propagation chain. The tracing results generated by this process clearly present the origin, transmission path, and potential impact of the anomaly, providing structured information and decision-making basis for subsequently identifying responsible parties, developing targeted remediation strategies, and implementing proactive risk management.

[0059] Specifically, based on the detection results, abnormal process nodes are marked, and bidirectional tracing is performed along the network of low-voltage construction process nodes to locate the source process of the abnormality and subsequent affecting processes, including:

[0060] Based on the process deviation early warning signal, the corresponding abnormal process node is marked in the low-voltage construction process node network;

[0061] Starting from the abnormal process node, trace back along the inter-process dependency relationship to locate the upstream process node in all the preceding dependencies where the actual construction data or construction timestamp is abnormal, and locate the node that no longer has abnormal preceding dependencies as the abnormal source process.

[0062] Starting from the abnormal process node, trace forward along the inter-process dependency relationship to locate all downstream process nodes that may be affected by the abnormality in the subsequent dependencies, and locate them as the subsequent affected processes.

[0063] The abnormal process nodes, the source process of the abnormality, and the subsequent affected processes are associated to construct a traceability result containing the complete abnormality propagation path.

[0064] First, based on the process deviation early warning signals generated by the construction data verification module, one or more abnormal process nodes are identified and marked in the low-voltage electrical construction process node network. This marking operation allows nodes with problems or deviations in the low-voltage electrical construction process node network to be explicitly identified, establishing a clear starting point for subsequent in-depth analysis.

[0065] Subsequently, a reverse tracing process is initiated. Starting from the aforementioned marked abnormal process nodes, the process traces back along the defined dependencies between process nodes to all their preceding dependent processes. During this process, the system checks the historical construction data and construction timestamp records of each upstream process node to locate nodes that also have abnormal actual construction data or construction timestamps. This tracing will continue until a node is found that, although it is marked as abnormal, none of its preceding dependent process nodes have abnormal records. This node is then identified as the abnormal source process of this anomaly event.

[0066] Simultaneously, a forward tracing process is initiated. Starting with the aforementioned marked anomalous process node, the investigation proceeds along the defined dependencies between process nodes, tracing towards all subsequent dependent processes. This process aims to assess the potential chain reaction triggered by the current anomaly. The system will locate all downstream process nodes that are directly or indirectly dependent on the anomalous node in the process flow. All located downstream nodes are identified as subsequently affected processes, indicating a potential risk that their construction quality or progress may be compromised due to the upstream anomaly.

[0067] Finally, the key nodes involved in this tracing, including the initially detected abnormal process nodes, the source process of the abnormality located through reverse tracing, and all subsequent affected processes located through forward tracing, are correlated and integrated. Specifically, based on the dependency links in the network of low-voltage electrical construction process nodes, a clear tracing result containing the complete abnormality propagation path can be constructed. This tracing result not only reveals the entire process of the abnormality from its source, through intermediate links, to its final detection, but also clarifies its potential future impact, thus providing structured decision support for a comprehensive understanding of the problem's causes, implementation of precise rectification, and risk warning.

[0068] The parameter dynamic optimization module 14 is used to dynamically optimize the process deviation warning threshold, generate an adaptive monitoring strategy, and update the weak current construction process node network based on the process deviation warning signal and the abnormal traceability result.

[0069] Because factors such as the construction environment, technological level, and personnel operation are dynamically changing, pre-set static technical parameter thresholds and monitoring strategies may not always accurately reflect actual project risks. This could lead to overly sensitive warnings resulting in redundant alarms, or delayed warnings missing the optimal intervention opportunity. Therefore, this parameter dynamic optimization module aims to continuously learn and analyze historical process deviation warning signals and their final traceability results to dynamically adjust the triggering conditions of various warning signals. This generates adaptive monitoring strategies that adapt to specific project stages and construction conditions, and feeds the optimized adaptive monitoring strategies back to the process node network. This enables the construction quality monitoring system to have the ability to self-evolve and continuously improve, thereby continuously enhancing the accuracy of warnings and the predictability of management.

[0070] Specifically, based on the process deviation early warning signal and the anomaly tracing results, the process deviation early warning threshold is dynamically optimized, an adaptive monitoring strategy is generated, and updated to the low-voltage construction process node network, including:

[0071] Aggregate historical process deviation warning signals and associated anomaly tracing results, and statistically analyze the frequency of each type of warning signal ultimately evolving into the anomaly source process in the tracing results;

[0072] Analyze the correlation strength between the deviation amplitude in the process deviation early warning signal and the severity of anomalies in the traceability results;

[0073] Based on the frequency and correlation strength, the trigger thresholds for various process deviation early warning signals are adjusted to generate an adaptive monitoring strategy that includes updated thresholds and early warning logic.

[0074] The adaptive monitoring strategy is bound to the corresponding process node in the low-voltage construction process node network to complete the update.

[0075] First, historically accumulated process deviation early warning signals and their associated anomaly tracing results obtained through analysis by the anomaly tracing and location module are aggregated. Based on this aggregated data, the frequency with which each type of early warning signal ultimately evolves into the anomaly source process in the tracing results is statistically analyzed. This frequency reflects the probability that a specific type of early warning signal indicates a true root cause problem.

[0076] Secondly, the correlation between the deviation magnitude recorded in the process deviation early warning signal and the severity of the anomaly in the anomaly tracing results is analyzed. By calculating the statistical correlation between the deviation magnitude and the severity of the anomaly, the rationality of the current early warning threshold setting can be assessed, and it can be determined whether a small deviation indicates a significant risk or whether a large deviation has a limited actual impact.

[0077] Specifically, the correlation strength between the deviation amplitude in the process deviation early warning signal and the severity of the anomaly in the traceability results is analyzed, including:

[0078] Based on the anomaly propagation path in the anomaly tracing results, the number of subsequent affected process nodes affected by each anomaly source process is counted, which is used as the corresponding anomaly severity.

[0079] For each abnormal event, extract the deviation magnitude recorded in the latest process deviation warning signal generated before the abnormal traceability result is produced in the source process of the abnormality.

[0080] The extracted deviation magnitude and the severity of the anomaly are paired to form a deviation magnitude-anomaly severity paired dataset;

[0081] Based on the aforementioned deviation magnitude-abnormality severity paired dataset, the Pearson correlation coefficient calculation method is used to calculate the linear correlation coefficient between the deviation magnitude and the abnormality severity, and the absolute value is taken as the correlation strength.

[0082] First, based on the identified anomaly propagation paths in the anomaly tracing results, for each located anomaly source process, the number of all subsequent affected process nodes directly or indirectly impacted by it is counted. This number of subsequent affected process nodes serves as a quantitative indicator to characterize the severity of the corresponding anomaly event. The more downstream process nodes affected, the wider the scope of the anomaly, and the higher the potential quality risks and repair costs; therefore, the greater the anomaly severity value.

[0083] Secondly, for each identified anomaly, the latest process deviation warning signal generated by the source process before triggering this anomaly tracing analysis is extracted. The recorded deviation magnitude is then read from this process deviation warning signal. This deviation magnitude reflects the immediate degree to which the parameters monitored by the system deviate from the preset threshold before the anomaly is finally confirmed.

[0084] Furthermore, the deviation magnitude value corresponding to each anomalous event extracted in the above steps is paired with its calculated anomaly severity value to form a deviation magnitude-anomaly severity paired dataset consisting of multiple data pairs. This deviation magnitude-anomaly severity paired dataset establishes a large number of empirical correlations between the characteristics of early warning signals and the final impact of anomalies.

[0085] Finally, based on the constructed bias magnitude-anomaly severity paired dataset, the Pearson correlation coefficient method is used to quantify the linear correlation between the two. The calculation process follows these steps: First, calculate the average of all bias magnitude data and the average of all anomaly severity data in the paired dataset. Second, for each pair of data in the dataset, calculate the difference between its bias magnitude and the average bias magnitude, and the difference between its anomaly severity and the average anomaly severity. Third, multiply the two differences for each pair of data and sum all the products to obtain the covariance numerator. Fourth, calculate the sum of squares of the differences between the bias magnitude data and its average, and the sum of squares of the differences between the anomaly severity data and its average. Fifth, divide the obtained covariance numerator by the square root of the product of the two sums of squares; the result is the Pearson correlation coefficient. The Pearson correlation coefficient ranges in the interval [-1, 1]. Taking the absolute value of the Pearson correlation coefficient is defined as the strength of the association between bias magnitude and anomaly severity.

[0086] The correlation strength is a value between 0 and 1. The closer the value is to 1, the stronger the positive linear relationship between the deviation magnitude and the severity of the anomaly. In other words, the greater the deviation in the warning signal, the more severe the anomaly's impact tends to be. This correlation strength can provide crucial data for the subsequent dynamic optimization of the warning threshold.

[0087] Furthermore, based on the frequency of each warning signal type and the correlation strength between its deviation amplitude and the severity of the anomaly calculated above, the original trigger thresholds of various process deviation warning signals are systematically adjusted. The adjustment logic is that for warning types that frequently transform into real source anomalies and whose deviation amplitude and severity are closely correlated, the monitoring sensitivity should be appropriately increased; for warning types with low transformation frequency or weak correlation, the monitoring conditions can be appropriately relaxed to reduce unnecessary interference alarms.

[0088] Based on the frequency and correlation strength, the trigger thresholds for various process deviation early warning signals are adjusted to generate an adaptive monitoring strategy that includes updated thresholds and early warning logic, including:

[0089] Based on the frequency and correlation strength of each warning signal type, a corresponding comprehensive adjustment coefficient is calculated, wherein the comprehensive adjustment coefficient is positively correlated with the frequency and the correlation strength;

[0090] Determine whether the correlation strength exceeds a preset threshold value and determine the adjustment direction. When the correlation strength exceeds the threshold value, the adjustment direction is to improve the early warning sensitivity.

[0091] Based on the adjustment direction, the comprehensive adjustment coefficient is multiplied by the preset benchmark adjustment range to obtain the threshold adjustment amount for the current warning signal type;

[0092] Based on the threshold adjustment amount and direction, the original process deviation warning signal trigger threshold is corrected to generate an updated warning threshold;

[0093] All warning signal types are bound to their corresponding updated warning thresholds and integrated to form an adaptive monitoring strategy.

[0094] First, for each type of process deviation warning signal, a comprehensive adjustment coefficient is calculated based on its statistically obtained frequency and the calculated correlation strength. This comprehensive adjustment coefficient is the product of the frequency value and the correlation strength value, and its value ranges between 0 and 1. The comprehensive adjustment coefficient comprehensively reflects the likelihood that this type of warning signal indicates a real root cause problem, as well as the strength of the correlation between the magnitude of the warning deviation and the final severity of the anomaly. A larger comprehensive adjustment coefficient indicates a higher necessity and potential benefit in optimizing the threshold of this type of warning signal.

[0095] Secondly, it is determined whether the correlation strength corresponding to the current warning signal type exceeds a preset threshold. This preset threshold is a pre-set value, such as 0.5, and is set based on the conventional statistical standards for classifying correlation strength and the engineering judgment of construction risk tolerance. This threshold is used to distinguish whether there is a sufficiently significant and reliable statistical correlation between the deviation amplitude of the warning signal and the abnormal consequences.

[0096] When the correlation strength exceeds the critical value, it indicates that there is a strong correlation between the deviation magnitude and the severity of the anomaly. The adjustment direction should be to improve the early warning sensitivity, that is, to tighten the early warning threshold in order to capture deviations that may cause serious problems earlier and more sensitively. Conversely, the adjustment direction should be to reduce the early warning sensitivity, that is, to moderately relax the early warning threshold in order to reduce unnecessary interference alarms.

[0097] Then, based on the adjustment direction determined in the previous step, the calculated comprehensive adjustment coefficient is multiplied by the preset baseline adjustment range to obtain the specific threshold adjustment amount for the current warning signal type. The baseline adjustment range is a pre-set adjustment ratio reference value based on the warning parameter category; for example, it can be set to 10% for physical parameter thresholds, 5% for signal parameter thresholds, and 15% for functional parameter thresholds. The magnitude of the threshold adjustment amount is determined jointly by the comprehensive adjustment coefficient and the baseline adjustment range.

[0098] Furthermore, based on the calculated threshold adjustment amount and the determined adjustment direction, the original process deviation warning signal trigger threshold is mathematically corrected.

[0099] Specifically, based on the threshold adjustment amount and direction, the original process deviation warning signal trigger threshold is corrected to generate an updated warning threshold, including:

[0100] Based on the threshold adjustment amount, a correction coefficient corresponding to the adjustment direction is determined, wherein the correction coefficient is used to convert the threshold adjustment amount into an adjustment ratio relative to the original process deviation warning signal trigger threshold;

[0101] The original process deviation warning signal trigger threshold is multiplied by the correction coefficient to generate a preliminary updated threshold;

[0102] The initial update threshold is compared with the preset minimum allowable threshold. If the initial update threshold is lower than the minimum allowable threshold, the minimum allowable threshold is used as the updated warning threshold; otherwise, the initial update threshold is used as the updated warning threshold.

[0103] First, based on the calculated threshold adjustment amount, a correction coefficient corresponding to the adjustment direction is determined. This correction coefficient is used to convert the absolute value of the threshold adjustment amount into an adjustment ratio relative to the original process deviation warning signal trigger threshold. Specifically, when the adjustment direction is to increase warning sensitivity, the direction factor is set to -1; when the adjustment direction is to decrease warning sensitivity, the direction factor is set to +1. Correction coefficient = 1 + direction factor × (threshold adjustment amount / original threshold). Through this calculation, the correction coefficient directly reflects the relative extent to which the threshold needs to be tightened or loosened. Second, the original process deviation warning signal trigger threshold is multiplied by the calculated correction coefficient to generate the initial updated threshold.

[0104] Therefore, to ensure that the updated preliminary threshold does not exceed the bottom line of engineering safety and quality, it is necessary to compare the generated preliminary threshold with the preset minimum allowable threshold. The minimum allowable threshold is determined based on the lower limit of the engineering threshold of technical parameters, representing the minimum standard that cannot be exceeded to ensure construction quality and system function.

[0105] When the initial update threshold is lower than the minimum allowable threshold, it indicates that the result of the automatic optimization calculation is too lenient and may jeopardize the project quality. In this case, the system will discard the initial result and directly use the minimum allowable threshold as the final updated warning threshold. If the initial update threshold is not lower than the minimum allowable threshold, it indicates that the optimization result is within a safe range, and the initial update threshold can be directly adopted as the updated warning threshold. This step introduces a safety constraint mechanism into the dynamic optimization process, ensuring that the adaptive adjustment of the monitoring strategy always remains within the boundaries allowed by the technical specifications.

[0106] Finally, all the types of early warning signals that need to be monitored in the system are bound to their corresponding updated early warning thresholds to form a complete and executable adaptive monitoring strategy. This adaptive monitoring strategy enables the dynamic evolution and personalized customization of early warning rules based on actual construction performance data.

[0107] Finally, the generated adaptive monitoring strategy is bound and integrated with the corresponding process nodes in the low-voltage construction process node network, completing the update of the monitoring rules in the low-voltage construction process node network. This enables the entire construction management system to continuously evolve based on actual operating data and feedback. Its monitoring behavior has the dynamic adaptability to self-optimize based on historical performance, thereby continuously improving the accuracy and efficiency of quality risk identification in long-term operation.

[0108] The report generation and solidification module 15 is used to generate construction management reports that include abnormal process nodes, abnormal types, traceability paths and related early warnings, and to solidify them with digital signatures and trusted timestamps.

[0109] Since the construction management process involves multiple responsible parties, and the resulting construction management report may serve as a key basis for quality assessment, project acceptance, and even subsequent dispute resolution, it is crucial to ensure the completeness, authenticity, immutability, and authoritativeness of the content and the time of its creation of the construction management report.

[0110] The purpose of this report generation and solidification module is to integrate the structured data and conclusions obtained from the analyses of the aforementioned modules into a formal construction management report with a standardized format and comprehensive content. By applying a digital signature and a trusted timestamp to the report, the electronic report is given evidentiary value, making its content and generation time verifiable and auditable afterward. This provides evidence for closed-loop management of the construction process, quality traceability, and liability determination.

[0111] Specifically, a construction management report is generated that includes abnormal process nodes, abnormal types, traceability paths, and associated early warnings, and is digitally signed and time-stamped, including:

[0112] The system summarizes the identifiers of the abnormal work nodes, their corresponding abnormal types, abnormal propagation paths, and historical process deviation warning signals associated with the abnormal work nodes to generate a structured construction management report.

[0113] Perform a hash calculation on the construction management report to generate a corresponding report summary;

[0114] The report digest is signed using a pre-set encrypted private key to generate digital signature data.

[0115] Submit the report summary to a trusted timestamp service to obtain a timestamp certificate containing authoritative time information;

[0116] The construction management report, digital signature data, and timestamp credentials are combined and packaged to output the final, solidified construction management report.

[0117] First, the key data generated by the anomaly tracing and location module and the parameter dynamic optimization module are summarized, including the identifier of the abnormal work process node, the corresponding anomaly type, the anomaly propagation path, and the historical process deviation warning signals associated with the abnormal work process node. This information is then organized and integrated to generate a complete and clearly structured digital construction management report.

[0118] Secondly, a cryptographic hash calculation is performed on the generated construction management report. Specifically, a hash calculation is a one-way function operation that maps report content of arbitrary length to a fixed-length and unique digital fingerprint, i.e., the report digest. Any slight modification to the original report content will cause the calculated digest to change, thus ensuring that the integrity of the report content is easily verifiable.

[0119] Then, using a pre-configured and securely stored encrypted private key during the system deployment phase, a digital signature operation is performed on the report digest calculated above. Specifically, the signature operation utilizes asymmetric encryption technology to generate a digital signature data that uniquely corresponds to the report digest and the pre-configured encrypted private key. This digital signature data proves that the report was issued by the entity holding the corresponding private key, and that its content has not been tampered with since issuance, thereby ensuring the authenticity of the report's source and the integrity of its content.

[0120] Secondly, the report summary is submitted to a trusted timestamp service provided by a third-party organization. This trusted timestamp service signs the received report summary and adds time information, generating a trusted timestamp certificate. This timestamp certificate independently proves that the report summary existed before its precise indicated time, providing proof of the report's existence.

[0121] Finally, the original construction management report, the generated digital signature data, and the obtained trusted timestamp credentials are combined and packaged to output a solidified final construction management report. This final construction management report integrates complete process data, identity authentication information, and authoritative time evidence, forming an electronic document with evidentiary value, long-term preservation, and easy verification. This provides a solid and reliable technical basis for construction quality traceability, liability determination, and project auditing.

[0122] In summary, the embodiments of this application have at least the following technical effects:

[0123] This invention digitally models the subway low-voltage electrical construction process, constructing a structured network of construction process nodes. This enables the systematic integration and visualization of construction information, eliminating the information silos present in traditional management. By collecting construction data and timestamps in real time and verifying continuity and consistency, proactive monitoring and deviation warnings of the construction process are achieved, improving the timeliness and accuracy of quality control.

[0124] Meanwhile, by utilizing a bidirectional traceability mechanism based on a process node network, the source of anomalies can be quickly located and its potential impact on downstream processes can be accurately assessed. This solves the problems of broken traceability chains and unclear scope of impact in traditional methods, improving the efficiency of problem investigation and responsibility determination. By analyzing the statistical patterns of historical early warning and traceability data, dynamic optimization of early warning thresholds and adaptive adjustment of monitoring strategies are achieved, enabling the management system to continuously learn and evolve, thereby continuously improving the accuracy of early warnings and the predictability of management.

[0125] Ultimately, by generating construction management reports that integrate complete traceability information and are solidified through digital signatures and trusted timestamps, authoritative electronic credentials are provided for project quality management and accountability. This promotes a fundamental shift in construction management towards intelligence, refinement, and traceability, demonstrating technological progress and positive effects in ensuring project quality, improving construction efficiency, clarifying the responsibilities of all parties, and supporting technological upgrades in the industry.

[0126] Example 2: This embodiment of the invention also provides a digital-based construction management method, applied to the digital-based construction management system described in Example 1, including:

[0127] First, based on the subway engineering design specifications, a network of low-voltage electrical construction process nodes is established, in which each process node includes the process type, technical parameter thresholds, and inter-process dependencies.

[0128] Specifically, based on subway engineering design specifications, a network of low-voltage electrical construction process nodes is established, including:

[0129] Extract the cable laying, equipment installation, and connector fabrication procedures defined in the subway low-voltage construction drawings and process documents, and determine them as the procedure types in the procedure node set;

[0130] According to the construction acceptance standards, corresponding technical parameter thresholds are set for each process type, including physical parameter thresholds, signal parameter thresholds, and functional parameter thresholds.

[0131] Analyze the construction process sequence between each process, establish the temporal and logical dependencies between process nodes, and form a network topology with directionality and hierarchy;

[0132] Based on the process types, the technical parameter thresholds corresponding to each process type, and the dependencies between process nodes, a network of process nodes for low-voltage electrical construction is constructed.

[0133] Secondly, collect actual construction data and construction timestamps for each process node, and verify continuity and consistency. When a deviation is detected, generate a process deviation warning signal.

[0134] Specifically, actual construction data and timestamps for each process node are collected, and continuity and consistency are verified. When a deviation is detected, a process deviation warning signal is generated, including:

[0135] Collect actual construction data and corresponding construction timestamps for each process node, wherein the actual construction data includes measured parameter values ​​of the same type as the technical parameter threshold;

[0136] Based on the temporal dependencies in the inter-process dependencies, verify whether the construction timestamp of each process node is later than the latest completion timestamp of all preceding dependent process nodes;

[0137] Based on the logical dependencies in the inter-process dependencies and the technical parameter thresholds, verify whether the measured parameter values ​​in the actual construction data of each process node meet the corresponding threshold range;

[0138] When time sequence continuity verification or data logic consistency verification fails, a process deviation warning signal is generated, including:

[0139] When timing continuity verification fails, a timing deviation warning signal is generated, which includes the timing deviation type, time offset, and identifier of the involved process node.

[0140] When data logic consistency verification fails, a parameter deviation warning signal is generated, which includes the parameter deviation type, the deviation of the measured parameter value from the technical parameter threshold, and the identifier of the involved process node.

[0141] The timing deviation warning signal and / or the parameter deviation warning signal are bound and encapsulated with the corresponding construction timestamp and process node information to form the process deviation warning signal.

[0142] Furthermore, based on the detection results, abnormal process nodes are marked, and bidirectional tracing is performed along the network of low-voltage construction process nodes to locate the source process of the abnormality and subsequent affected processes.

[0143] Specifically, based on the detection results, abnormal process nodes are marked, and bidirectional tracing is performed along the network of low-voltage construction process nodes to locate the source process of the abnormality and subsequent affecting processes, including:

[0144] Based on the process deviation early warning signal, the corresponding abnormal process node is marked in the low-voltage construction process node network;

[0145] Starting from the abnormal process node, trace back along the inter-process dependency relationship to locate the upstream process node in all the preceding dependencies where the actual construction data or construction timestamp is abnormal, and locate the node that no longer has abnormal preceding dependencies as the abnormal source process.

[0146] Starting from the abnormal process node, trace forward along the inter-process dependency relationship to locate all downstream process nodes that may be affected by the abnormality in the subsequent dependencies, and locate them as the subsequent affected processes.

[0147] The abnormal process nodes, the source process of the abnormality, and the subsequent affected processes are associated to construct a traceability result containing the complete abnormality propagation path.

[0148] Furthermore, based on the process deviation early warning signal and the anomaly tracing results, the process deviation early warning threshold is dynamically optimized, an adaptive monitoring strategy is generated, and updated to the weak current construction process node network.

[0149] Specifically, based on the process deviation early warning signal and the anomaly tracing results, the process deviation early warning threshold is dynamically optimized, an adaptive monitoring strategy is generated, and updated to the low-voltage construction process node network, including:

[0150] Aggregate historical process deviation warning signals and associated anomaly tracing results, and statistically analyze the frequency of each type of warning signal ultimately evolving into the anomaly source process in the tracing results;

[0151] Analyze the correlation strength between the deviation amplitude in the process deviation early warning signal and the severity of anomalies in the traceability results;

[0152] Based on the frequency and correlation strength, the trigger thresholds for various process deviation early warning signals are adjusted to generate an adaptive monitoring strategy that includes updated thresholds and early warning logic.

[0153] The adaptive monitoring strategy is bound to the corresponding process node in the low-voltage construction process node network to complete the update.

[0154] Specifically, the correlation strength between the deviation amplitude in the process deviation early warning signal and the severity of the anomaly in the traceability results is analyzed, including:

[0155] Based on the anomaly propagation path in the anomaly tracing results, the number of subsequent affected process nodes affected by each anomaly source process is counted, which is used as the corresponding anomaly severity.

[0156] For each abnormal event, extract the deviation magnitude recorded in the latest process deviation warning signal generated before the abnormal traceability result is produced in the source process of the abnormality.

[0157] The extracted deviation magnitude and the severity of the anomaly are paired to form a deviation magnitude-anomaly severity paired dataset;

[0158] Based on the aforementioned deviation magnitude-abnormality severity paired dataset, the Pearson correlation coefficient calculation method is used to calculate the linear correlation coefficient between the deviation magnitude and the abnormality severity, and the absolute value is taken as the correlation strength.

[0159] Specifically, based on the frequency and correlation strength, the trigger thresholds for various process deviation early warning signals are adjusted to generate an adaptive monitoring strategy that includes updated thresholds and early warning logic, including:

[0160] Based on the frequency and correlation strength of each warning signal type, a corresponding comprehensive adjustment coefficient is calculated, wherein the comprehensive adjustment coefficient is positively correlated with the frequency and the correlation strength;

[0161] Determine whether the correlation strength exceeds a preset threshold value and determine the adjustment direction. When the correlation strength exceeds the threshold value, the adjustment direction is to improve the early warning sensitivity.

[0162] Based on the adjustment direction, the comprehensive adjustment coefficient is multiplied by the preset benchmark adjustment range to obtain the threshold adjustment amount for the current warning signal type;

[0163] Based on the threshold adjustment amount and direction, the original process deviation warning signal trigger threshold is corrected to generate an updated warning threshold;

[0164] All warning signal types are bound to their corresponding updated warning thresholds and integrated to form an adaptive monitoring strategy.

[0165] Specifically, based on the threshold adjustment amount and direction, the original process deviation warning signal trigger threshold is corrected to generate an updated warning threshold, including:

[0166] Based on the threshold adjustment amount, a correction coefficient corresponding to the adjustment direction is determined, wherein the correction coefficient is used to convert the threshold adjustment amount into an adjustment ratio relative to the original process deviation warning signal trigger threshold;

[0167] The original process deviation warning signal trigger threshold is multiplied by the correction coefficient to generate a preliminary updated threshold;

[0168] The initial update threshold is compared with the preset minimum allowable threshold. If the initial update threshold is lower than the minimum allowable threshold, the minimum allowable threshold is used as the updated warning threshold; otherwise, the initial update threshold is used as the updated warning threshold.

[0169] Finally, a construction management report is generated, which includes abnormal process nodes, abnormal types, traceability paths, and associated early warnings, and is then digitally signed and secured with a trusted timestamp.

[0170] Specifically, a construction management report is generated that includes abnormal process nodes, abnormal types, traceability paths, and associated early warnings, and is digitally signed and time-stamped, including:

[0171] The system summarizes the identifiers of the abnormal work nodes, their corresponding abnormal types, abnormal propagation paths, and historical process deviation warning signals associated with the abnormal work nodes to generate a structured construction management report.

[0172] Perform a hash calculation on the construction management report to generate a corresponding report summary;

[0173] The report digest is signed using a pre-set encrypted private key to generate digital signature data.

[0174] Submit the report summary to a trusted timestamp service to obtain a timestamp certificate containing authoritative time information;

[0175] The construction management report, digital signature data, and timestamp credentials are combined and packaged to output the final, solidified construction management report.

Claims

1. A digitally-based construction management system, characterized in that: The system includes: The process network construction module is used to establish a network of low-voltage construction process nodes according to the subway engineering design specifications. Each process node includes the process type, technical parameter thresholds, and inter-process dependencies. The construction data verification module is used to collect actual construction data and construction timestamps for each process node, and verify continuity and consistency. When a deviation is detected, a process deviation warning signal is generated. The anomaly tracing and location module is used to mark abnormal process nodes based on the detection results, and to perform bidirectional tracing along the network of low-voltage construction process nodes to locate the source process of the anomaly and subsequent affected processes. The parameter dynamic optimization module is used to dynamically optimize the process deviation warning threshold based on the process deviation warning signal and the abnormality tracing result, generate an adaptive monitoring strategy, and update it to the weak current construction process node network. The report generation and consolidation module is used to generate construction management reports that include abnormal process nodes, abnormal types, traceability paths and related early warnings, and to solidify them with digital signatures and trusted timestamps.

2. The digital-based construction management system according to claim 1, characterized in that, According to the subway engineering design specifications, a network of low-voltage electrical construction process nodes is established, including: Extract the cable laying, equipment installation, and connector fabrication procedures defined in the subway low-voltage construction drawings and process documents, and determine them as the procedure types in the procedure node set; According to the construction acceptance standards, corresponding technical parameter thresholds are set for each process type, including physical parameter thresholds, signal parameter thresholds, and functional parameter thresholds. Analyze the construction process sequence between each process, establish the temporal and logical dependencies between process nodes, and form a network topology with directionality and hierarchy; Based on the process types, the technical parameter thresholds corresponding to each process type, and the dependencies between process nodes, a network of process nodes for low-voltage electrical construction is constructed.

3. The digital-based construction management system according to claim 1, characterized in that, Collect actual construction data and timestamps for each process node, and verify continuity and consistency. When a deviation is detected, generate a process deviation warning signal, including: Collect actual construction data and corresponding construction timestamps for each process node, wherein the actual construction data includes measured parameter values ​​of the same type as the technical parameter threshold; Based on the temporal dependencies in the inter-process dependencies, verify whether the construction timestamp of each process node is later than the latest completion timestamp of all preceding dependent process nodes; Based on the logical dependencies in the inter-process dependencies and the technical parameter thresholds, verify whether the measured parameter values ​​in the actual construction data of each process node meet the corresponding threshold range; When time sequence continuity verification or data logic consistency verification fails, a process deviation warning signal is generated, including: When timing continuity verification fails, a timing deviation warning signal is generated, which includes the timing deviation type, time offset, and identifier of the involved process node. When data logic consistency verification fails, a parameter deviation warning signal is generated, which includes the parameter deviation type, the deviation of the measured parameter value from the technical parameter threshold, and the identifier of the involved process node. The timing deviation warning signal and / or the parameter deviation warning signal are bound and encapsulated with the corresponding construction timestamp and process node information to form the process deviation warning signal.

4. The digital-based construction management system according to claim 1, characterized in that, Based on the detection results, abnormal process nodes are marked, and bidirectional tracing is performed along the network of low-voltage construction process nodes to locate the source process of the abnormality and subsequent affecting processes, including: Based on the process deviation early warning signal, the corresponding abnormal process node is marked in the low-voltage construction process node network; Starting from the abnormal process node, trace back along the inter-process dependency relationship to locate the upstream process node in all the preceding dependencies where the actual construction data or construction timestamp is abnormal, and locate the node that no longer has abnormal preceding dependencies as the abnormal source process. Starting from the abnormal process node, trace forward along the process dependencies to locate all downstream process nodes that may be affected by the abnormality in subsequent dependencies, and locate them as subsequent affected processes. The abnormal process nodes, the source process of the abnormality, and the subsequent affected processes are associated to construct a traceability result containing the complete abnormality propagation path.

5. The digital-based construction management system according to claim 1, characterized in that, Based on the process deviation early warning signal and anomaly tracing results, the process deviation early warning threshold is dynamically optimized, an adaptive monitoring strategy is generated, and updated to the low-voltage construction process node network, including: Aggregate historical process deviation warning signals and associated anomaly tracing results, and statistically analyze the frequency of each type of warning signal ultimately evolving into the anomaly source process in the tracing results; Analyze the correlation strength between the deviation amplitude in the process deviation early warning signal and the severity of anomalies in the traceability results; Based on the frequency and correlation strength, the trigger thresholds for various process deviation early warning signals are adjusted to generate an adaptive monitoring strategy that includes updated thresholds and early warning logic. The adaptive monitoring strategy is bound to the corresponding process node in the low-voltage construction process node network to complete the update.

6. The digital-based construction management system according to claim 5, characterized in that, The correlation strength between the deviation amplitude in the process deviation warning signal and the severity of the anomaly in the traceability results is analyzed, including: Based on the anomaly propagation path in the anomaly tracing results, the number of subsequent affected process nodes affected by each anomaly source process is counted, which is used as the corresponding anomaly severity. For each abnormal event, extract the deviation magnitude recorded in the latest process deviation warning signal generated before the abnormal traceability result is produced in the source process of the abnormality. The extracted deviation magnitude and the severity of the anomaly are paired to form a deviation magnitude-anomaly severity paired dataset; Based on the aforementioned deviation magnitude-abnormality severity paired dataset, the Pearson correlation coefficient calculation method is used to calculate the linear correlation coefficient between the deviation magnitude and the abnormality severity, and the absolute value is taken as the correlation strength.

7. The digital-based construction management system according to claim 5, characterized in that, Based on the frequency and correlation strength, the trigger thresholds for various process deviation early warning signals are adjusted to generate an adaptive monitoring strategy that includes updated thresholds and early warning logic, including: Based on the frequency and correlation strength of each warning signal type, a corresponding comprehensive adjustment coefficient is calculated, wherein the comprehensive adjustment coefficient is positively correlated with the frequency and the correlation strength; Determine whether the correlation strength exceeds a preset threshold value and determine the adjustment direction. When the correlation strength exceeds the threshold value, the adjustment direction is to improve the early warning sensitivity. Based on the adjustment direction, the comprehensive adjustment coefficient is multiplied by the preset benchmark adjustment range to obtain the threshold adjustment amount for the current warning signal type; Based on the threshold adjustment amount and direction, the original process deviation warning signal trigger threshold is corrected to generate an updated warning threshold; All warning signal types are bound to their corresponding updated warning thresholds and integrated to form an adaptive monitoring strategy.

8. The digital-based construction management system according to claim 7, characterized in that, Based on the threshold adjustment amount and direction, the original process deviation early warning signal trigger threshold is corrected to generate an updated early warning threshold, including: Based on the threshold adjustment amount, a correction coefficient corresponding to the adjustment direction is determined, wherein the correction coefficient is used to convert the threshold adjustment amount into an adjustment ratio relative to the original process deviation warning signal trigger threshold; The original process deviation warning signal trigger threshold is multiplied by the correction coefficient to generate a preliminary updated threshold; The initial update threshold is compared with the preset minimum allowable threshold. If the initial update threshold is lower than the minimum allowable threshold, the minimum allowable threshold is used as the updated warning threshold; otherwise, the initial update threshold is used as the updated warning threshold.

9. The digital-based construction management system according to claim 1, characterized in that, Generate a construction management report that includes abnormal process nodes, abnormal types, traceability paths, and associated early warnings, and solidify it with digital signatures and trusted timestamps, including: The system summarizes the identifiers of the abnormal work nodes, their corresponding abnormal types, abnormal propagation paths, and historical process deviation warning signals associated with the abnormal work nodes to generate a structured construction management report. Perform a hash calculation on the construction management report to generate a corresponding report summary; The report digest is signed using a pre-set encryption private key to generate digital signature data. Submit the report summary to a trusted timestamp service to obtain a timestamp certificate containing authoritative time information; The construction management report, digital signature data, and timestamp credentials are combined and packaged to output the final, solidified construction management report.

10. A digitally-based construction management method, characterized in that: The method is applied to the digital-based construction management system according to any one of claims 1-9, and the method includes: According to the subway engineering design specifications, a network of low-voltage electrical construction process nodes is established, in which each process node includes process type, technical parameter thresholds and inter-process dependencies. Collect actual construction data and construction timestamps for each process node, and verify continuity and consistency. When a deviation is detected, generate a process deviation warning signal. Based on the test results, abnormal process nodes are marked, and bidirectional tracing is performed along the network of low-voltage construction process nodes to locate the source process of the abnormality and subsequent affected processes. Based on the process deviation early warning signal and the anomaly tracing results, the process deviation early warning threshold is dynamically optimized, an adaptive monitoring strategy is generated, and updated to the weak current construction process node network. Generate a construction management report that includes abnormal process nodes, abnormal types, traceability paths, and associated warnings, and solidify it with digital signatures and trusted timestamps.

Images