Work procedure collaborative operation early warning system based on tunnel large machine matched construction

Abstract

The invention belongs to the technical field of tunnel matching construction early warning, and particularly relates to a process collaborative operation early warning system based on tunnel large-machine matching construction, which performs multi-source cross validation on process events uploaded in tunnel large-machine matching construction by fusing space positioning information and real-time operation state signals of corresponding responsible equipment of the process events. According to the method, credible judgment on the authenticity of an original process event is realized, meanwhile, the process event passing cross verification is taken as a node, a process dependency network is constructed based on the sequence dependency and parallel cooperation relationship between the processes, and the time sequence coupling logic of multiple processes is explicitly described; and after the collaborative deviation is identified, recursively deducing a conduction path of the delay under the constraint of a sequence chain and parallel by taking a deviation node as a starting point. Therefore, early warning is upgraded from single-point overtime judgment to global evaluation based on full-link influence, a prospective basis is provided for active scheduling, and meanwhile, early warning triggering depends on actual influence diffusion, so that early warning timeliness and accuracy are improved.

Description

Technical Field

[0001] This invention belongs to the field of early warning technology for tunnel construction operations, specifically relating to an early warning system for collaborative operations based on tunnel construction using large machinery. Background Technology

[0002] In tunnel drilling and blasting construction, large-scale mechanized equipment is typically used in conjunction with other equipment, such as three-arm drilling rigs, wet spraying robots, and hydraulic inverted arch trestle bridges, forming a cyclical operation chain of "excavation-muck removal-support-lining". These devices need to work together efficiently during construction; otherwise, it will lead to prolonged process cycle time and delayed construction progress.

[0003] Currently, common process management methods in tunnel construction typically rely on IoT interfaces on equipment or manual reporting to obtain process events, and trigger timeout alarms based on preset time thresholds. However, their warning logic generally simplifies each process into independent time nodes, judging solely by whether the actual duration of a single process exceeds the planned value, without considering the sequential dependencies and parallel constraints between processes. Therefore, when a process is delayed, its chain reaction effect on downstream processes and its potential impact on the overall construction path are difficult to identify. This isolated and static warning mechanism leads to fragmented risk perception and a lack of a global perspective. It cannot provide a reliable basis for proactive intervention and is prone to issuing warnings prematurely before deviations have had a substantial impact, reducing on-site trust in the warnings.

[0004] Furthermore, automatically collected equipment signals may be distorted due to communication interruptions or sensor malfunctions; manually entered process completion information may contain false alarms. Existing systems often directly use these raw data for analysis, lacking verification of whether process events actually occurred, which can easily lead to false deviation identification or missed detection of real anomalies, further weakening the reliability of the early warning system. Summary of the Invention

[0005] To solve the above-mentioned technical problems, or at least partially solve them, the present invention provides an early warning system for collaborative operation of tunnel construction using large-scale machinery.

[0006] The objective of this invention can be achieved through the following technical solution: a collaborative operation early warning system for tunnel construction supported by large machinery, comprising: a process event acquisition module: acquiring and uploading process events from the tunnel construction management platform, including event name, associated work face, responsible equipment and timestamp.

[0007] Cross-validation module: Uses spatial positioning information and status signals of responsible equipment to cross-validate process events.

[0008] Process Dependency Building Module: Using cross-validated process events as nodes and the order and parallel relationships between processes as edges, a process dependency network is constructed.

[0009] Deviation identification module: It uses the process dependency network to locate process events and compares them with the sequential baseline interval and parallel time difference tolerance based on the statistics of historical process events to identify collaborative deviations.

[0010] Transmission and deduction module: When identifying collaborative deviations, starting from the deviation event, it traverses subsequent processes according to the process dependency network, deduces the transmission path of the deviation in the sequential chain and parallel constraints, forms the influence transmission chain, and analyzes the types of nodes traversed by the influence transmission chain.

[0011] Early warning generation module: Generates early warning information based on the nodes and types along the influence transmission chain.

[0012] Combining all the above technical solutions, the positive effects of this invention are as follows: 1. By integrating the spatial positioning information of the corresponding responsible equipment with the real-time operating status signal of the process events uploaded during the construction of tunnel machinery, this invention performs multi-source cross-verification, thereby achieving a reliable judgment on the authenticity of the original process events. It can filter out false events caused by communication interruption, sensor abnormality, human error, or timestamp misalignment to the greatest extent possible, thus improving the reliability of the early warning system.

[0013] 2. This invention uses verified process events as nodes and constructs a process dependency network based on the sequential dependency and parallel collaboration relationship between processes to explicitly characterize the temporal coupling logic of multiple processes. After identifying collaboration deviations, starting from the deviation node, the propagation path of delay under sequential chain and parallel constraints is recursively deduced, so that the early warning is upgraded from a single-point timeout judgment to a global assessment based on the impact of the entire link, providing a forward-looking basis for proactive scheduling. At the same time, since the early warning trigger depends on the actual impact spread, it effectively avoids premature alarms and improves timeliness and accuracy. Attached Figure Description

[0014] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0015] Figure 1 This is a schematic diagram of the system module connections of the present invention.

[0016] Figure 2 This is a flowchart illustrating the implementation of the cross-validation module in this invention.

[0017] Figure 3 This is a flowchart illustrating the implementation of identifying cooperative deviations in this invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] See Figure 1 As shown, the present invention provides a collaborative operation early warning system for tunnel construction using large-scale machinery, comprising a process event acquisition module, a cross-validation module, a process dependency construction module, a deviation identification module, a propagation and deduction module, and an early warning generation module connected in sequence.

[0020] The process event acquisition module is used to collect and upload process events from the tunnel machinery supporting construction management platform, which include event name, associated working face, responsible equipment and timestamp.

[0021] In tunnel construction using large machinery, a variety of large-scale mechanized equipment are typically involved, and their operation is highly time-dependent. Throughout the construction cycle, various operational nodes naturally generate process events, such as the completion of face excavation, the start of initial support, the completion of invert concrete pouring, and the placement of the lining trolley. These events are the basic data units for achieving progress awareness and therefore need to be captured and recorded.

[0022] To obtain the above events, the construction process monitors the operating status (such as start / stop signals) and spatial location of the equipment through the Internet of Things or manual monitoring. When the monitoring data meets the criteria for completion of a process, the corresponding process event will be generated or manually reported and uploaded to the tunnel construction management platform.

[0023] Specifically, the collected process events include not only the event name and timestamp, but also the associated working face and responsible equipment. The associated working face refers to the specific construction space where the process event occurs, which is usually defined by the mileage interval and working face number in tunnel engineering, and is used to anchor the process event to the physical construction area.

[0024] Responsible equipment refers to mechanized equipment that directly performs the required procedure. For example, a three-arm drilling rig is responsible for excavation preparation before drilling and blasting, a wet shotcrete robot is responsible for initial concrete spraying for support, and a hydraulic invert arch trestle is responsible for invert arch construction. Responsible equipment information is used to establish a mapping relationship between procedures and equipment.

[0025] The cross-validation module is used to cross-validate process events using spatial positioning information and status signals of the responsible equipment.

[0026] Uploaded process events may contain false information due to reasons such as equipment communication interruptions or human error. If such unreliable events are directly used for subsequent collaborative deviation identification, it will lead to misjudgment of deviations or even trigger invalid warnings. Therefore, it is necessary to verify the collected process events.

[0027] To achieve the above objectives, this invention considers that any real process event is inevitably accompanied by the objective physical behavior of the responsible equipment within a specific working surface, and this behavior can be characterized by obtainable objective signals such as the spatial position and operating status of the equipment. Therefore, by comparing the information contained in the process event with the perceived data from the equipment side, the authenticity of the event can be effectively verified.

[0028] See Figure 2 As shown, in a specific embodiment of the present invention, the cross-validation module contains the following: after receiving a process event, based on the timestamp of the event, a list of all devices in operation at the time corresponding to that timestamp is obtained from the controller interface of the tunnel construction equipment.

[0029] The responsible equipment for the process event is matched with a list of equipment in operation. If the responsible equipment is in the list, the actual spatial location of the responsible equipment within a preset time window before and after the event timestamp is further extracted from the equipment positioning system.

[0030] The actual spatial location is compared with the spatial range of the work surface associated with the process event. If the actual location of the responsible equipment falls within the spatial range of the work surface, the process event is determined to have passed cross-validation; otherwise, it has not passed cross-validation.

[0031] The process dependency construction module is used to construct a process dependency network with cross-validated process events as nodes and the order and parallel relationships between processes as edges.

[0032] The process events retained after cross-validation are real and valid events, which can accurately reflect the actual construction progress on site. Given that there are strict temporal logic and collaborative constraints between different processes during the construction of tunnel machinery, for example, the construction of the invert arch must be started only after the initial support is completed, which reflects the sequential dependency relationship, while the left and right tunnel faces need to be excavated synchronously to control the deformation of the surrounding rock, which reflects the parallel constraint relationship. These collaborative relationships constitute the framework of the construction logic.

[0033] Therefore, based on the process events that have passed the cross-validation, a process dependency network containing sequential and parallel edges can be explicitly constructed, thereby mapping the collaborative structure of the physical construction process in the digital space and providing topological support for subsequent deviation identification and deviation propagation deduction.

[0034] As a preferred implementation, the process dependency network is constructed as follows: using process events as network nodes and the sequential and parallel relationships between processes as directed edges, the initial process dependency relationship is constructed.

[0035] Assign attributes to each sequential relationship edge and record the predecessor and successor nodes of the connected nodes.

[0036] Assign attributes to each parallel relationship edge and record the node groups involved.

[0037] The collaborative deviation identification module is used to locate process events using the process dependency network and compare them with the sequential benchmark interval and parallel time difference tolerance based on historical process event statistics to identify collaborative deviations.

[0038] Deviation analysis can only be performed based on the actual time interval between a process event and its adjacent or parallel process events after the process event has been cross-validated and its authenticity has been ensured. The prerequisite for such deviation identification is the establishment of a reliable judgment benchmark.

[0039] Considering that the construction process has strong repeatability and regularity under stable working conditions, this invention extracts the time benchmark between processes from historically verified standard construction cycle segments without coordinated deviation as the judgment benchmark for coordinated deviation of processes.

[0040] In one optional implementation, the sequential baseline interval and parallel time difference tolerance obtained based on the statistics of historical process events are implemented as follows: from the historical process event sequence that has passed cross-validation, normal construction cycle segments that are determined to have no collaborative deviation are selected.

[0041] For each sequential relationship edge in the process dependency network, statistical analysis is performed on the distribution of the time interval between the end time of the preceding node event and the start time of the succeeding node event in all normal construction cycle segments, forming a time interval sample set for that dependency edge.

[0042] For each parallel relationship edge involved in the process dependency network, the difference between the start times of the node groups in all normal construction cycle segments is counted to form the parallel time difference sample set of the parallel relationship.

[0043] Frequency histograms are plotted for the time interval sample set of sequential relationship edges and the parallel time difference sample set of parallel relationship edges. The continuous interval with the highest frequency in the histogram is identified as the main peak interval, which represents the most frequent process connection interval during construction. The adjacent low-frequency intervals are expanded to both sides of the main peak interval until the cumulative frequency coverage reaches the critical value (such as 90%). The finally formed extended interval is used as the normal fluctuation benchmark range.

[0044] For sequential relationships, this reference range serves as the sequential reference interval.

[0045] For parallel relationships, this baseline range serves as the parallel time difference tolerance interval.

[0046] See Figure 3 As shown, further, identifying collaborative deviations includes the following: receiving verified process event streams in real time, and locating the corresponding node in the process dependency network when a new process event arrives.

[0047] Find all incoming edges of the node. If there is an incoming edge with a sequential relationship, obtain the most recent process event of its predecessor node, calculate the actual time interval between the two, and compare it with the sequential reference interval interval corresponding to the sequential relationship edge. If the actual interval is not within the reference interval interval, it is determined that there is a deviation in the sequential relationship.

[0048] Find the nodes associated with the parallel relationship edge to which the node belongs, check the start time of the most recent event of the associated node, calculate the actual time difference between the start time of the event of the node and the start time of the most recent event of the associated node, and determine that the parallel relationship is out of balance when the actual time difference exceeds the parallel time difference tolerance range corresponding to the parallel relationship edge.

[0049] Process events that indicate sequential deviations or parallel misalignments are marked as deviation events.

[0050] The propagation deduction module is used to, when identifying collaborative deviations, take the deviation event as the starting point, traverse subsequent processes according to the process dependency network, deduce the propagation path of the deviation in the sequential chain and parallel constraints, form an influence propagation chain, and analyze the types of nodes traversed by the influence propagation chain.

[0051] When a process deviates from its coordinated operation, due to the strong interdependence between processes, this deviation often propagates downwards along the network, triggering a chain reaction and even lengthening the overall cycle time. To achieve a comprehensive risk assessment, it is necessary to extrapolate the scope and transmission effects of the deviation.

[0052] The following describes the propagation process with reference to specific embodiments: S1. Taking the network node corresponding to the deviation event as the root node, and combining its actual occurrence time with the required execution time of the process, the completion time is predicted. The predicted completion time is compared with the original planned start time of the subsequent node to obtain the predicted time interval between the two. Furthermore, the predicted time interval is compared with the reference interval of the corresponding sequential edge in the process dependency network to identify the cooperative deviation. If the predicted interval is not within the reference interval, it is determined that the subsequent node is in a cooperative deviation state due to the upstream deviation propagation. The connection between the root node and the subsequent node is recorded as the initial segment of the propagation path.

[0053] S2. For each new node included in the transmission path, determine whether it belongs to a parallel relationship node group. If it does, since the nodes in the group satisfy the synchronization constraint in time, associate the previous node of the current transmission path with all member nodes in the parallel relationship node group, and generate a branch path from the previous node of the current transmission path to all member nodes in the parallel relationship node group to represent the influence channel that the deviation may affect other processes in the same group through the parallel constraint.

[0054] S3. For each newly visited node, recursively execute the above time prediction, cooperative deviation identification and parallel path expansion operations, and continue to traverse downwards along its sequential edges until the current node does not trigger a new cooperative deviation or reaches the end of the network.

[0055] S4. During the entire traversal, record the propagation path from the root node to each affected end node. Each complete path is defined as an influence propagation chain.

[0056] The execution time of the procedures and the original planned start time of the nodes used in the above simulation are all planned parameters of the tunnel machinery construction. They can be obtained from the construction schedule and used as the initial input and comparison benchmark for deviation propagation simulation.

[0057] It should be noted that the deviation impact propagation simulation used in this invention does not mechanically add up the time delay of the root node. Instead, it considers that even if the upstream process is delayed, if the sequential baseline interval between it and the subsequent process is sufficiently large, the delay may be naturally absorbed and will not cause the actual start time of the subsequent process to deviate from the reasonable execution window. Therefore, this system performs deviation re-identification in each propagation step, reflecting the propagation condition judgment of step-by-step verification. Only when the delay truly encroaches on the normal window of the subsequent process is the impact considered to have been propagated, effectively avoiding false diffusion warnings caused by blindly accumulating delays and improving the engineering realism of the propagation simulation.

[0058] After completing the influence transmission simulation, given that different types of process nodes have different impacts on the construction progress, this invention further analyzes the type attributes of the nodes through which each influence transmission chain passes in order to quantitatively assess the engineering severity of the deviation propagation.

[0059] In one alternative implementation, the types of nodes in the transmission path are analyzed as follows: (1) Nodes that have sequential incoming edges and belong to at least one parallel node group are marked as dual-constraint nodes. Such nodes are subject to both sequential logic and parallel collaboration window constraints, and have very low adjustment freedom. Once affected by upstream deviation, they are very likely to cause local process blockages and may trigger chain imbalances within the parallel group. They are high-risk nodes in construction.

[0060] (2) Nodes with two or more responsible equipment for a process event are marked as multi-equipment collaborative nodes. The execution of such nodes depends on the spatial coordination and operation connection of multiple large equipment. Any delay or scheduling conflict of any equipment may cause the entire process to stop. They are more sensitive to deviations and more difficult to recover, and are also high-risk nodes.

[0061] In the above analysis of node types, the first type of dual-constraint node is based on the connection relationship of the process in the network topology, which belongs to the structural analysis at the process logic level.

[0062] The second type of multi-device collaborative node is based on the mapping relationship between the process and physical resources, that is, by parsing the number of responsible devices associated with the process event of the node, which belongs to the execution analysis at the resource allocation level.

[0063] Both methods identify high-risk nodes from two dimensions: process collaboration logic and equipment resource allocation.

[0064] The early warning generation module is used to generate early warning information based on the nodes and their types in the influence transmission chain.

[0065] Specifically, the early warning generation process is as follows: The number and type of nodes along the influence transmission chain are statistically analyzed. An early warning is triggered when any of the following conditions are met: a) The number of nodes along the influence transmission chain exceeds a preset node threshold, reflecting that the deviation influence has crossed multiple process stages, forming a wide-area diffusion effect. The node threshold can be determined based on the number of processes included in the tunnel machinery supporting construction cycle. When there are many processes and the process is long in the construction cycle, the node threshold is increased accordingly; when there are few processes and the process is short in the construction cycle, the node threshold is decreased accordingly. For example, more than half of the total number of processes in the construction cycle can be taken, such as using 60% of the total number of processes as the node threshold.

[0066] b) If, before the length of the transmission chain reaches the threshold of the number of nodes, there are already dual-constraint nodes or multi-device collaborative nodes among the nodes it passes through, it indicates that the collaborative deviation has affected high-risk nodes and is very likely to trigger a chain reaction or safety risk.

[0067] Condition a focuses on the breadth of impact; condition b focuses on the depth of impact. Both conditions characterize the severity of risk from different dimensions, and reaching a critical state in either dimension provides early warning value.

[0068] After detecting a deviation in process coordination, this invention performs forward impact deduction based on the process dependency network to generate an impact transmission chain representing the deviation propagation path. Based on the number and type of nodes involved in the chain, it comprehensively determines whether to trigger an early warning. Compared with traditional methods that only issue an alarm when the actual completion time of a process exceeds the planned threshold, the early warning mechanism of this solution is essentially a forward-looking risk perception based on causal logic. On the one hand, it can filter out minor disturbances that can be absorbed due to sufficient buffering, reducing early warning fatigue. On the other hand, it intervenes before the deviation causes actual delays, reserving a time window for scheduling adjustments.

[0069] In further innovative implementation, when an early warning is triggered, a sequence of nodes along the impact transmission chain is generated, and the type of each node and its position in the chain are marked in the sequence, so that managers can understand the risk evolution path and enhance their trust in the early warning judgment.

[0070] Furthermore, the early warning generation module also includes spatial priority analysis of the generated impact transmission chain, specifically as follows: Obtain the associated working surfaces of all nodes in the impact transmission chain corresponding to the process events, and compare them with the associated working surfaces of the root node: If all affected nodes in the impact transmission chain are located on the same working surface as the root node, then the impact transmission chain is marked as a working surface impact chain, and given high display priority in the generated early warning sequence. High-priority early warnings can trigger pop-up placement, audio-visual reminders, or automatic push to the terminal of the person in charge of the working face, so as to achieve rapid on-site processing. This is because processes within the same working face share limited spatial channels, working face resources, and large equipment access windows. Deviation impacts have strong coupling and immediate interference, which can easily cause local blockages or even loop interruptions, requiring immediate on-site handling.

[0071] If there are affected nodes on different working faces than the root node in the impact transmission chain, the impact transmission chain is marked as a cross-working face impact chain and given normal display priority in the generated warning sequence. This is because cross-working face impacts usually spread through indirect paths such as equipment scheduling and personnel allocation, and their urgency is relatively low. There may be buffer or isolation mechanisms, such as independent working faces or backup equipment, which can be coordinated and handled later.

[0072] By analyzing the spatial priority of the consistency of associated work surfaces, this invention classifies the influence transmission chain into two categories: the same work surface and cross work surfaces. Based on this, the priority of early warning display is adjusted to ensure that dispatchers pay priority to high coupling deviations that are most likely to cause the current work surface to stagnate, and avoid being distracted by cross-area influences with low urgency.

[0073] In summary, the present invention implements early warning for collaborative deviation by first determining whether to trigger an early warning based on the number and type of nodes in the influence transmission chain; when an early warning is triggered, it generates early warning information containing node sequences and type labels; furthermore, it performs spatial priority analysis on the generated early warning information based on the working face information of each node, distinguishing and displaying the influence chain within the same working face from the influence chain across working faces. This reflects a leap from coarse-grained timeout alarms to fine-grained, interpretable, and spatially aware collaborative risk early warning, which is conducive to improving the guiding value of early warning information for on-site scheduling decisions.

[0074] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0075] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0076] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0077] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0078] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A collaborative operation early warning system for tunnel construction using large-scale machinery, characterized in that, include: Work process event acquisition module: Collects and uploads work process events from the tunnel machinery supporting construction management platform, including event name, associated work face, responsible equipment and timestamp; Cross-validation module: Uses spatial positioning information and status signals of responsible equipment to cross-validate process events; Process Dependency Building Module: Using cross-validated process events as nodes and the sequential and parallel relationships between processes as edges, a process dependency network is constructed. Deviation identification module: It uses the process dependency network to locate process events and compares them with the sequential baseline interval and parallel time difference tolerance based on the statistics of historical process events to identify collaborative deviations; Transmission and deduction module: When identifying collaborative deviation, starting from the deviation event, it traverses subsequent processes according to the process dependency network, deduces the transmission path of the deviation in the sequential chain and parallel constraints, forms the influence transmission chain, and analyzes the types of nodes traversed by the influence transmission chain; Early warning generation module: Generates early warning information based on the nodes and types along the influence transmission chain.

2. The early warning system for collaborative operations in tunnel construction based on large-scale tunnel machinery as described in claim 1, characterized in that: The specific contents of the cross-validation module are as follows: Upon receiving a process event, based on the timestamp of the event, obtain a list of all devices that are in operation at that timestamp from the controller interface of the tunnel construction equipment; Match the responsible equipment for the process event with a list of equipment in operation. If the responsible equipment is in the list, further extract the actual spatial location of the responsible equipment within a preset time window before and after the event timestamp from the equipment positioning system. The actual spatial location is compared with the spatial range of the work surface associated with the process event. If the actual location of the responsible equipment falls within the spatial range of the work surface, the process event is determined to have passed cross-validation; otherwise, it has not passed cross-validation.

3. The early warning system for collaborative operations in tunnel construction based on large-scale tunnel machinery as described in claim 1, characterized in that: The process-dependent network is described in the following construction process: Using process events as network nodes and the sequential and parallel relationships between processes as directed edges, the initial process dependency relationships are constructed. Assign attributes to each sequential relationship edge to record the predecessor and successor nodes of the connected nodes; Assign attributes to each parallel relationship edge and record the node groups involved.

4. The early warning system for collaborative operations in tunnel construction based on large-scale tunnel machinery as described in claim 1, characterized in that: The sequential baseline interval and parallel time difference tolerance based on historical process event statistics are determined as follows: From the historical process event sequences that have passed cross-validation, normal construction cycle segments that are determined to have no collaborative deviations are selected; For each sequential relationship edge in the process dependency network, the distribution of the time interval between the end time of the preceding node event and the start time of the succeeding node event is statistically analyzed in all normal construction cycle segments, forming a time interval sample set for that sequential relationship edge; For each parallel relationship edge involved in the process dependency network, the difference between the start times of the node groups in all normal construction cycle segments is counted to form the parallel time difference sample set of the parallel relationship. Frequency histograms were plotted for the time interval sample set of sequential relation edges and the parallel time difference sample set of parallel relation edges, and the normal fluctuation reference range was extracted from them. For sequential relationships, this reference range serves as the sequential reference interval. For parallel relationships, this baseline range serves as the parallel time difference tolerance interval.

5. The early warning system for collaborative operations in tunnel construction based on large-scale tunnel machinery as described in claim 4, characterized in that: The identification of collaborative deviations includes the following: Receive verified process event streams in real time, and locate the corresponding node in the process dependency network when a new process event arrives; Find all incoming edges of the node. If there is an incoming edge with a sequential relationship, obtain the most recent process event of its predecessor node, calculate the actual time interval between the two, and compare it with the sequential reference interval interval corresponding to the sequential relationship edge. If the actual interval is not within the reference interval interval, it is determined that the sequential relationship has deviated. Find the nodes associated with the parallel relationship edge to which the node belongs, check the start time of the most recent event of the associated node, calculate the actual time difference between the start time of the event of the node and the start time of the most recent event of the associated node, and determine that the parallel relationship is out of balance when the actual time difference exceeds the parallel time difference tolerance range corresponding to the parallel relationship edge. Process events that indicate sequential deviations or parallel misalignments are marked as deviation events.

6. The early warning system for collaborative operations in tunnel construction based on large-scale tunnel machinery as described in claim 5, characterized in that: The propagation deduction module performs the following operations: Taking the network node corresponding to the deviation event as the root node, the completion time is predicted by combining its actual occurrence time with the required execution time of the process. The predicted completion time is compared with the original planned start time of the subsequent node to obtain the predicted time interval between the two. Furthermore, the predicted time interval is compared with the sequential reference interval of the corresponding sequential edge in the process dependency network to identify collaborative deviation. If collaborative deviation is identified, the connection between the root node and the subsequent node is recorded as the initial segment of the transmission path. For each new node included in the propagation path, determine whether it belongs to a parallel relation node group; If it belongs to the category, then generate a branch path from the previous node of the current propagation path to all member nodes in the parallel relation node group; For each newly visited node, the above time prediction, cooperative deviation identification and parallel path expansion operations are recursively executed, and the nodes are continuously traversed downwards along their sequential edges until the current node does not trigger a new cooperative deviation or reaches the end of the network. Throughout the traversal, the propagation path from the root node to each affected end node is recorded, and each complete path is defined as an influence propagation chain.

7. The early warning system for collaborative operations in tunnel construction based on large-scale tunnel machinery as described in claim 1, characterized in that: The analysis of the influence of the types of nodes traversed by the transmission chain includes the following: A node that has both sequential incoming edges and belongs to at least one parallel relation node group is marked as a double-constraint node. Nodes with two or more responsible devices for a process event are marked as multi-device collaborative nodes.

8. The early warning system for collaborative operations in tunnel construction based on large-scale tunnel machinery as described in claim 7, characterized in that: The warning generation module is specifically used for: The system counts the number and type of nodes along the influence propagation chain, and triggers an alert when any of the following conditions are met: a) The number of nodes traversed by the influence transmission chain exceeds a preset threshold; b) The influence of the transmission chain is that before the length reaches the threshold of the number of nodes, there are already dual-constraint nodes or multi-device collaborative nodes among the nodes it passes through.

9. The early warning system for collaborative operations in tunnel construction based on large-scale tunnel machinery as described in claim 8, characterized in that: When an early warning is triggered, the early warning generation module generates a sequence of nodes that affect the transmission chain and marks the type of each node and its position in the chain in the sequence.

10. The early warning system for collaborative operations in tunnel construction based on large-scale tunnel machinery as described in claim 1, characterized in that: The early warning generation module also includes spatial priority analysis of the generated impact transmission chain: Obtain the associated working surfaces of the process events that affect all nodes in the transmission chain, and compare them with the associated working surfaces of the root node: If all affected nodes in the influence transmission chain are located on the same working surface as the root node, then the influence transmission chain is marked as a working surface influence chain and given high display priority in the generated warning sequence; If there are affected nodes on different working surfaces than the root node in the influence propagation chain, then the influence propagation chain is marked as a cross-working surface influence chain and given normal display priority in the generated warning sequence.

Images