An operation management system and method for tunnel construction equipment
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA TIESIJU CIVIL ENGINEERING GROUP CO LTD
- Filing Date
- 2026-04-07
- Publication Date
- 2026-07-24
Smart Images
Figure CN122453124A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of digital management technology for tunnel construction, and in particular to an operation management system and method for tunnel construction equipment. Background Technology
[0002] Tunnel engineering is developing towards longer distances, larger cross-sections, and deeper burial depths. The construction environment is enclosed and complex, with dense equipment, placing extremely high demands on schedule, safety, and cost control. While technologies such as the Internet of Things and big data are driving the digital upgrade of construction management, adapting them to the unique challenges of tunnel environments remains a challenge.
[0003] The operation and management of core equipment such as tunnel boring machines and loaders suffer from problems such as difficulty in data collection, delays in scheduling, and insufficient risk warnings: signal obstruction leads to opaque location and status data, and there is a lack of basis for equipment utilization and cost accounting; manual inspection is inefficient, and the supply of key equipment is unbalanced; hidden dangers such as overtime operation and abnormal fuel consumption are difficult to detect in a timely manner, which can easily lead to failures and safety accidents.
[0004] Current solutions mainly fall into three categories: manual management relies on on-site records, which are prone to errors and omissions, making them unsuitable for large-scale construction; location monitoring solutions have limited dimensions, suffer from poor signal strength within tunnels, and lack key data such as fuel consumption and status; single-function monitoring tools present fragmented data, cannot be integrated with dispatching, and result in delayed early warning responses. Existing technologies generally suffer from incomplete data collection, unsystematic analysis, lack of functional coordination, and high reliance on manual intervention, making it difficult to meet the intelligent management needs of tunnel engineering. Summary of the Invention
[0005] In view of the above problems, an operation and management system and method for tunnel construction equipment are proposed to overcome or at least partially solve the above problems, specifically: An operation and management system for tunnel construction equipment, comprising: The monitoring device is deployed on the construction equipment inside the tunnel to collect real-time data on the operating status, fuel consumption, and location of the construction equipment. It uses an anti-interference data transmission module to transmit the operating status, fuel consumption, and location data to the data processing module. The anti-interference data transmission module includes an inertial measurement unit, an ad hoc network communication unit, a hybrid backhaul unit, a tunnel signal dynamic sensing engine, and a multi-link intelligent collaborative controller. The inertial measurement unit is used to maintain the continuity of position data through dead reckoning when satellite positioning signals fail. The ad hoc network communication unit is used to form an ad hoc network with adjacent monitoring devices and transmit data through multi-hop relays. The hybrid backhaul unit is used to form a hybrid backhaul link with a wireless network through a leaky coaxial cable. The tunnel signal dynamic sensing engine is used to collect multi-link signal quality parameters of the locations of each monitoring device in real time and construct a dynamic signal heat map of the entire tunnel. The multi-link intelligent collaborative controller is used to allocate transmission links for data of different priorities according to the dynamic signal heat map and optimize the ad hoc network topology in real time. The data processing module is used to integrate and analyze the received operating status data, fuel consumption data, and location data to generate equipment operation indicators. The early warning module is used to compare real-time acquired equipment operation indicators with user-defined early warning rules, and automatically generate and issue early warning information when the early warning conditions are met. The visual interface includes a map module for displaying the real-time location of construction equipment and a chart module for displaying the changing trends of equipment operation indicators; The scheduling module is used to receive early warning information and support managers to conduct voice communication with operators of construction equipment in the tunnel to execute scheduling instructions based on the changing trends of operational indicators.
[0006] Optionally, the operating status data includes at least one of the following: average daily working time of the equipment, real-time operating status, operating rate, and working efficiency; the fuel consumption data includes at least one of the following: average daily fuel consumption of the equipment and real-time fuel consumption; and the location data includes at least one of the following: real-time location coordinates of the equipment and historical movement trajectory.
[0007] Optionally, the equipment operation indicators generated by the data processing module include at least one of the following: equipment utilization rate index, cost control index, and failure risk index.
[0008] Optionally, the early warning module includes a rule configuration interface and an early warning statistics dashboard. The rule configuration interface allows users to customize threshold conditions for different equipment operation indicators to form early warning rules. The early warning statistics dashboard is used to classify and display the triggered early warning information according to the early warning type, associated equipment, and associated processes.
[0009] Optionally, the visual interface also includes an equipment management view, which displays the equipment number, equipment type, maintenance manager, and current working status of all registered construction equipment in a list format. It also supports managers to query, filter, add, and delete construction equipment by equipment number or equipment name.
[0010] Optionally, the device management view also includes a device status indicator that is linked in real time with the monitoring device. The device status indicator is used to distinguish and display the current working status as online, offline, standby, or faulty on the visual interface.
[0011] Optionally, the tunnel signal dynamic perception engine has a built-in signal quality acquisition unit. The signal quality acquisition unit is used to continuously collect satellite signal strength, self-organizing network link signal-to-noise ratio, and leakage coaxial signal attenuation value, generate a three-dimensional signal fingerprint containing location, time, and signal quality, and upload it to the edge computing node in the tunnel. The edge computing node constructs and updates the dynamic signal heat map of the entire tunnel according to a preset frequency based on the signal fingerprint.
[0012] Optionally, the multi-link intelligent collaborative controller divides the data to be transmitted into four priorities: control commands, early warning information, status data, and historical logs; and divides the data into signal-good areas, signal-weak areas, and signal-blind areas according to the dynamic signal heat map, so as to match the corresponding transmission links for data of different priorities.
[0013] Optionally, the multi-link intelligent collaborative controller is also used to calculate the adaptive topology of the ad hoc network in real time based on the real-time movement trajectory and dynamic signal heat map of the construction equipment, and automatically adjust the transmission power and relay path of the ad hoc network communication unit.
[0014] An operation and management method for tunnel construction equipment includes: The monitoring device collects real-time operating status data, fuel consumption data, and location data of the construction equipment, and uses an anti-interference data transmission module to transmit the operating status data, fuel consumption data, and location data to the data processing module. The data processing module integrates and analyzes the received operating status data, fuel consumption data, and location data to generate equipment operation indicators. The system acquires equipment operation indicators in real time through the early warning module, compares these indicators with user-defined early warning rules, and automatically generates and issues early warning information when the early warning conditions are met. The real-time location of construction equipment and the changing trends of equipment operation indicators are displayed through a visual interface. The scheduling module receives early warning information from the early warning module, and based on the changing trends of operational indicators, the management personnel conduct voice communication with the operators of construction equipment in the tunnel to execute scheduling instructions.
[0015] This invention provides a tunnel construction equipment operation management system and method. The system collects real-time operating status, fuel consumption, and location data through monitoring devices and utilizes an anti-interference data transmission module to ensure stable data transmission in the tunnel environment. A data processing module integrates and analyzes data to generate equipment operation indicators, quantifying equipment utilization, costs, and risks. An early warning module compares indicators in real-time based on custom rules, automatically triggering warnings when anomalies occur, achieving proactive risk warning. A visual interface uses a map module to accurately locate equipment positions and a chart module to dynamically present indicator trends, making equipment distribution and operational status clear at a glance. After receiving warnings, the dispatch module supports one-click voice communication for management personnel, issuing dispatch instructions based on indicator trends. This system improves data transparency and dispatch response efficiency, shortens fault handling and resource allocation time, reduces the risk of missed or false alarms, and provides tunnel engineering with a low-cost, highly adaptable intelligent operation management method. Attached Figure Description
[0016] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the present invention will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 This is a flowchart of an operation and management method for tunnel construction equipment provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the login interface of a smart construction site big data platform provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the main interface of a smart construction site big data platform provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the data dashboard interface of a smart construction site big data platform provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of a device information interface provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of a hardware management interface provided in an embodiment of the present invention; Figure 7a This is a schematic diagram of an early warning center dashboard provided in an embodiment of the present invention; Figure 7b This is a schematic diagram of a rule configuration interface provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of a device data interface provided in an embodiment of the present invention; Figure 9 This is a schematic diagram of a data analysis interface provided in an embodiment of the present invention. Detailed Implementation
[0018] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0019] This invention provides an operation and management system for tunnel construction equipment, which may specifically include: The monitoring device is deployed on the construction equipment inside the tunnel to collect real-time data on the operating status, fuel consumption, and location of the construction equipment. It uses an anti-interference data transmission module to transmit the operating status, fuel consumption, and location data to the data processing module. The anti-interference data transmission module includes an inertial measurement unit, an ad hoc network communication unit, a hybrid backhaul unit, a tunnel signal dynamic sensing engine, and a multi-link intelligent collaborative controller. The inertial measurement unit is used to maintain the continuity of position data through dead reckoning when satellite positioning signals fail. The ad hoc network communication unit is used to form an ad hoc network with adjacent monitoring devices and transmit data through multi-hop relays. The hybrid backhaul unit is used to form a hybrid backhaul link with a wireless network through a leaky coaxial cable. The tunnel signal dynamic sensing engine is used to collect multi-link signal quality parameters of the locations of each monitoring device in real time and construct a dynamic signal heat map of the entire tunnel. The multi-link intelligent collaborative controller is used to allocate transmission links for data of different priorities according to the dynamic signal heat map and optimize the ad hoc network topology in real time. The data processing module is used to integrate and analyze the received operating status data, fuel consumption data, and location data to generate equipment operation indicators. The early warning module is used to compare real-time acquired equipment operation indicators with user-defined early warning rules, and automatically generate and issue early warning information when the early warning conditions are met. The visual interface includes a map module for displaying the real-time location of construction equipment and a chart module for displaying the changing trends of equipment operation indicators; The scheduling module is used to receive early warning information and support managers to conduct voice communication with operators of construction equipment in the tunnel to execute scheduling instructions based on the changing trends of operational indicators.
[0020] The operational status data includes at least one of the following: average daily operating time, real-time operating status, utilization rate, and operating efficiency; fuel consumption data includes at least one of the following: average daily fuel consumption and real-time fuel consumption; location data includes at least one of the following: real-time location coordinates and historical movement trajectory. The equipment operation indicators generated by the data processing module include at least one of the following: equipment utilization rate index, cost control index, and failure risk index.
[0021] In one specific embodiment of the present invention, the operation and management system for tunnel construction equipment can be implemented in the form of a "smart construction site big data platform." This platform integrates three core management functions: project, equipment, and data, and is accessible through a login interface (such as...). Figure 2 After completing user authentication (as shown), you will enter the platform's main interface (as shown). Figure 3 (As shown). The platform deploys monitoring devices on each construction machine inside the tunnel. These devices connect to the equipment controller's local area network bus and the added fuel consumption sensors and inertial navigation units via physical interfaces, continuously collecting operational status data, fuel consumption data, and location data. Operational status data includes at least the equipment's average daily working time, real-time operating status (tunneling, standby, shutdown, fault, idle, etc.), operating rate, and work efficiency. Fuel consumption data includes at least the average daily fuel consumption and real-time fuel consumption. Location data includes at least real-time location coordinates and historical movement trajectories composed of continuous coordinate sequences. The monitoring device has a built-in anti-interference data transmission module. This module uses a dedicated communication protocol adapted to the tunnel's enclosed environment, maintaining a stable connection with the platform server even in tunnel sections with severe satellite signal obstruction, and transmitting the collected data frames back to the data processing module in encrypted format in real time.
[0022] In one embodiment of the present invention, the anti-interference data transmission module may consist of an inertial measurement unit, an ad hoc network communication unit, a hybrid backhaul unit, a tunnel signal dynamic sensing engine, and a multi-link intelligent collaborative controller.
[0023] Specifically, the inertial measurement unit (IMU) is connected to the built-in satellite positioning module of the monitoring device, receiving the position information output by the satellite positioning module in real time. When the satellite positioning signal fails, the IMU initiates the dead reckoning process, collecting the device's acceleration and angular velocity information to calculate the device's displacement and attitude changes, thereby continuously generating the device's position information and maintaining the continuity of the position information until the satellite positioning signal is restored.
[0024] The ad hoc network communication unit has dual functions as a terminal node and a relay node. When the monitoring device is powered on, the ad hoc network communication unit automatically searches for other ad hoc network communication units in the vicinity, establishes connections with adjacent ad hoc network communication units, and jointly builds a distributed ad hoc network. When an ad hoc network communication unit cannot directly establish communication with the gateway at the tunnel entrance, the unit sends the information to be transmitted to an adjacent ad hoc network communication unit, which then acts as a relay node to continue forwarding the information. After multiple relays, the information is finally transmitted to the gateway at the tunnel entrance, realizing multi-hop relay transmission.
[0025] The hybrid backhaul unit integrates both a leaky coaxial cable communication interface and a wireless network communication interface. The leaky coaxial cable is laid along the tunnel, while the wireless network covers the tunnel's operating area. The hybrid backhaul unit establishes a connection with the leaky coaxial cable through the leaky coaxial cable communication interface and with the wireless network inside the tunnel through the wireless network communication interface. The two links operate in parallel, together forming a hybrid backhaul link. Information to be transmitted can be transmitted to the gateway at the tunnel entrance via either the leaky coaxial cable link or the wireless network link.
[0026] The tunnel signal dynamic sensing engine connects to the satellite positioning module, the self-organizing network communication unit, and the hybrid backhaul unit, respectively, to collect real-time satellite positioning signal quality parameters, self-organizing network link signal quality parameters, leaky coaxial cable link signal quality parameters, and wireless network link signal quality parameters at the locations of each monitoring device. The engine correlates the collected multi-link signal quality parameters with the location information and acquisition time information of the corresponding monitoring devices. After integrating the correlated information uploaded by all monitoring devices, it constructs a dynamic signal heat map of the entire tunnel, reflecting the signal quality distribution at different locations and times throughout the tunnel.
[0027] The multi-link intelligent collaborative controller connects to the tunnel signal dynamic sensing engine to acquire a real-time dynamic signal heatmap of the entire tunnel. Based on the priority of the information to be transmitted and the signal quality at the information transmission location, it allocates corresponding transmission links for information of different priorities. Simultaneously, based on the signal distribution reflected in the full tunnel dynamic signal heatmap, the multi-link intelligent collaborative controller adjusts the connection relationships and transmission parameters of each node in the ad hoc network in real time, optimizing the network's topology and improving its transmission reliability and efficiency.
[0028] Specifically, the tunnel signal dynamic perception engine has a built-in signal quality acquisition unit. The signal quality acquisition unit is used to continuously collect satellite signal strength, self-organizing network link signal-to-noise ratio, and leakage coaxial signal attenuation value, generate a three-dimensional signal fingerprint containing location, time, and signal quality, and upload it to the edge computing node in the tunnel. The edge computing node constructs and updates the dynamic signal heat map of the entire tunnel according to a preset frequency based on the signal fingerprint.
[0029] The multi-link intelligent collaborative controller categorizes data to be transmitted into four priority levels: control commands, early warning information, status data, and historical logs. It also classifies data into signal-good areas, signal-weak areas, and signal-blind areas based on a dynamic signal heatmap, matching corresponding transmission links to data of different priorities. Furthermore, the multi-link intelligent collaborative controller is used to calculate the adaptive topology of the ad hoc network in real time based on the real-time movement trajectory of construction equipment and the dynamic signal heatmap, automatically adjusting the transmission power and relay paths of the ad hoc network communication units.
[0030] More specifically, the signal quality acquisition unit built into the tunnel signal dynamic perception engine can use a microcontroller as the core control chip. The printed circuit board of the signal quality acquisition unit has three sets of independent interface circuits reserved, which establish point-to-point electrical connections with the satellite positioning module, self-organizing network communication unit and hybrid backhaul unit built into the monitoring device, respectively.
[0031] The signal quality acquisition unit and the satellite positioning module can be connected via a serial peripheral interface. The serial peripheral interface is configured as a host mode, with the clock frequency set to a preset value. It adopts a transmission format with a preset data bit length and a preset transmission order. The clock polarity and clock phase are also set to preset values. The signal quality acquisition unit sends a preset read command frame to the satellite positioning module, triggering the satellite positioning module to return a response frame containing satellite signal strength parameters. The byte at a preset position in the response frame is the hexadecimal representation of the satellite signal strength parameters.
[0032] The signal quality acquisition unit and the ad hoc network communication unit can be connected via a universal asynchronous transceiver interface. The universal asynchronous transceiver interface is configured in full-duplex mode, with the baud rate set to a preset value. It adopts a transmission format with a preset data bit length, a preset stop bit length, and no parity bit. After each link detection with a neighboring node is completed, the ad hoc network communication unit automatically sends a data packet containing the ad hoc network link signal-to-noise ratio (SNR) parameter to the signal quality acquisition unit. The byte at a preset position in the data packet is the data packet identifier, and the bytes at the remaining preset positions are the ad hoc network link SNR parameters.
[0033] The signal quality acquisition unit and the hybrid return unit can be connected via an analog input interface. The hybrid return unit integrates a signal attenuation detection circuit, which converts the signal attenuation value of the leaky coaxial cable into a DC voltage signal within a preset range. The analog input interface of the signal quality acquisition unit uses an analog-to-digital converter with a preset resolution to convert the input DC voltage signal into a corresponding digital quantity, thereby calculating the leaky coaxial signal attenuation value parameter.
[0034] The signal quality acquisition unit can be configured with a timer interrupt. The trigger period of the timer interrupt is consistent with the preset acquisition period. Each time the timer interrupt is triggered, the signal quality acquisition unit starts a parameter acquisition process. The parameter acquisition process is executed sequentially in the order of satellite signal strength, self-organizing network link signal-to-noise ratio, and leaky coaxial signal attenuation value. After acquiring each parameter, the signal quality acquisition unit verifies the validity of the parameter. If the parameter value is within the preset reasonable range, it is determined to be a valid parameter and retained. If the parameter value exceeds the reasonable range, it is determined to be an invalid parameter and discarded. The parameter is then reacquired. If multiple acquisition failures occur consecutively, the parameter is marked as missing. After completing the acquisition of each of the three types of signal quality parameters, the signal quality acquisition unit sends a data request command to the main control unit of the monitoring device through the internal bus. After receiving the request command, the main control unit returns the current time information and the real-time location information of the construction equipment to the signal quality acquisition unit. The time information uses a common timestamp format, and the location information uses longitude and latitude in a common geographic coordinate system.
[0035] The signal quality acquisition unit concatenates the three types of signal quality parameters—satellite signal strength, self-organizing network link signal-to-noise ratio, and leaky coaxial signal attenuation value—with the corresponding location information and time information in a preset byte order to generate a fixed-length three-dimensional signal fingerprint. In the three-dimensional signal fingerprint, the bytes at preset positions are time information, the bytes at preset positions are location information, and the bytes at the remaining preset positions are the three types of signal quality parameters.
[0036] The signal quality acquisition unit writes the generated three-dimensional signal fingerprint into the circular buffer of the local storage unit of the monitoring device. The local storage unit uses a non-volatile memory chip, and the size of the circular buffer can accommodate a preset number of three-dimensional signal fingerprints. When the buffer is full, the newly generated three-dimensional signal fingerprint automatically overwrites the earliest written three-dimensional signal fingerprint.
[0037] The signal quality acquisition unit is internally equipped with both an upload timer and a data volume counter. When the upload timer reaches a preset upload interval, or when the number of unuploaded 3D signal fingerprints in the buffer counted by the data volume counter reaches a preset threshold, the preset upload trigger conditions are met. The signal quality acquisition unit then sends an upload request to the hybrid backhaul unit. The hybrid backhaul unit, based on the currently available transmission links, uploads all unuploaded 3D signal fingerprints in the buffer to the edge computing nodes within the tunnel in batches. If the transmission link is interrupted during the upload process, the signal quality acquisition unit stops the upload operation and continues to store newly generated 3D signal fingerprints in the circular buffer. Once the transmission link is restored, it automatically resumes uploading the remaining 3D signal fingerprints from the point of interruption.
[0038] After receiving the 3D signal fingerprints uploaded from all monitoring devices, the edge computing node first parses and verifies the 3D signal fingerprints, discarding invalid data that fails verification. Then, it performs spatial and temporal aggregation processing on all valid data. The edge computing node pre-divides the entire tunnel space into several square spatial grid cells with equal side lengths. Each spatial grid cell corresponds to a unique grid index. Based on the position information in the 3D signal fingerprint, the edge computing node calculates the spatial grid cell index to which the signal fingerprint belongs. The average value of all signal quality parameters within the same spatial grid cell and the same time window is taken as the signal quality value of that spatial grid cell within that time window.
[0039] The edge computing node uses the signal quality values of all spatial grid cells as a basis and employs a linear interpolation algorithm to generate a continuous signal quality distribution surface, thereby constructing a dynamic signal heatmap of the entire tunnel. The edge computing node refreshes the dynamic signal heatmap of the entire tunnel at a preset update frequency. Each refresh replaces the historical signal quality values in the corresponding spatial grid cells with the signal quality values within the latest time window, recalculates the signal quality distribution surface, and updates the heatmap. The updated dynamic signal heatmap of the entire tunnel is stored in the local storage unit of the edge computing node. Simultaneously, the edge computing node broadcasts the updated dynamic signal heatmap to the multi-link intelligent collaborative controller of all online monitoring devices.
[0040] The multi-link intelligent collaborative controller can use a microcontroller from the same series as the signal quality acquisition unit as its core control chip. The read-only memory of the multi-link intelligent collaborative controller pre-programs a data priority classification rule file, which defines the correspondence between data type identifiers and data priorities. The first type of data type identifier corresponds to control commands with a priority of level one; the second type corresponds to warning information with a priority of level two; the third type corresponds to status data with a priority of level three; and the fourth type corresponds to historical logs with a priority of level four. When the multi-link intelligent collaborative controller receives data to be transmitted from the main control unit of the monitoring device, it first extracts the data type identifier from the header of the data to be transmitted. Based on the correspondence between the data type identifier and the data priority classification rule, it assigns the corresponding priority to the data to be transmitted and adds the priority information to the header of the data to be transmitted.
[0041] The multi-link intelligent collaborative controller receives real-time dynamic signal heatmaps of the entire tunnel pushed by edge computing nodes and stores the latest dynamic signal heatmaps in its internal random access memory. At preset time intervals, the multi-link intelligent collaborative controller obtains the real-time location information of the construction equipment from the main control unit of the monitoring device, calculates the corresponding spatial grid cell index based on this location information, and then matches the signal quality value of that spatial grid cell from the dynamic signal heatmap of the entire tunnel.
[0042] The multi-link intelligent collaborative controller's read-only memory pre-stores preset signal quality threshold ranges. A first threshold and a second threshold divide the signal quality values into three intervals: a signal quality value greater than the first threshold corresponds to a good signal zone; a signal quality value between the first and second thresholds corresponds to a weak signal zone; and a signal quality value less than the second threshold corresponds to a dead signal zone. Based on the signal area where the construction equipment is located and the priority of the data to be transmitted, the multi-link intelligent collaborative controller executes preset transmission link matching rules, assigning corresponding transmission links to data of different priorities. In a good signal zone, all priority data is allowed to be transmitted via a wireless network link; in a weak signal zone, only first-priority control commands and second-priority warning information are allowed to be transmitted via a leaky coaxial cable link, while third-priority status data and fourth-priority historical logs are temporarily stored in the local storage unit; in a dead signal zone, only first-priority control commands and second-priority warning information are allowed to be transmitted via a self-organizing network link using multi-hop relay, while third-priority status data and fourth-priority historical logs are temporarily stored in the local storage unit.
[0043] The multi-link intelligent collaborative controller simultaneously acquires real-time movement trajectory information of the construction equipment. This real-time movement trajectory information is generated by the main control unit of the monitoring device based on continuous location information, including all location points and movement directions of the construction equipment within a preset time period. Combining the real-time movement trajectory information of the construction equipment with the dynamic signal heat map of the entire tunnel, the multi-link intelligent collaborative controller predicts the future location changes of the construction equipment within a preset time period and the corresponding signal quality changes, thereby calculating the adaptive topology of the self-organizing network in real time.
[0044] The multi-link intelligent collaborative controller sends a transmit power adjustment command to the ad hoc network communication unit via a universal asynchronous transceiver interface. This command includes a target transmit power value. Upon receiving the command, the ad hoc network communication unit adjusts its transmit power to the target value. Simultaneously, based on the signal quality values of adjacent nodes in the dynamic signal heatmap of the entire tunnel, the multi-link intelligent collaborative controller calculates the total signal quality of all possible relay paths. It selects the path with the higher total signal quality as the adapted relay path, generates new routing table information, and sends it to the ad hoc network communication unit. After the ad hoc network communication unit updates its local routing table, subsequent ad hoc network data transmissions are forwarded according to the new routing table, enabling the ad hoc network topology to dynamically adapt to changes in signal quality within the tunnel and the movement of construction equipment.
[0045] The data processing module, deployed on the platform's cloud or a central server outside the tunnel, continuously receives massive data streams from all monitoring devices. This module first performs protocol parsing, outlier cleaning, and timestamp alignment on the raw data. Then, it fuses operating status, fuel consumption, and location data from multiple sources according to device identifiers and time dimensions. Based on this, the data processing module executes quantitative analysis algorithms to generate equipment operation indicators: these include an equipment utilization rate index calculated based on the ratio of actual operating time to planned operating time; a cost control index based on the deviation of unit output fuel consumption from historical benchmarks; and a fault risk index estimated in real-time using a logistic regression model based on characteristics such as engine coolant temperature, vibration amplitude, and fault code frequency. These indicators are updated every five minutes and stored in a time-series database, on a per-device basis. The data processing module also communicates bidirectionally with the visualization interface, pushing indicator data to the front-end rendering engine.
[0046] The visual interface is the core interaction layer of the smart construction site big data platform, presented on the command center's large screen and management personnel's mobile terminals using a browser / server architecture. The main body of the interface includes data dashboards (such as...). Figure 4As shown in the image, the map module uses a high-precision electronic map of the tunnel, marking the real-time geographical location of each construction device with different colored icons. The icon color dynamically changes according to the current working status of the device (e.g., green for on-site, gray for offline, yellow for standby, red for fault). Clicking the icon will display a floating layer showing the device number, maintenance manager, real-time fuel consumption, and cumulative working hours. The map module also supports historical movement trajectory playback: after the user selects the device and time period, the system retrieves the stored coordinate sequence and draws a trajectory line with speed color markings on the map. The chart module is integrated into the left and bottom areas of the dashboard, using a pie chart to display the percentage of on-site, offline, standby, and fault statuses of the equipment cluster, a radar chart to comprehensively evaluate the fuel efficiency, working time, safety score, and other dimensions of a single device, and a line chart to show the changing trends of operational indicators such as equipment utilization index and fault risk index over the past seven and thirty days. Users can zoom in and out of the time axis using a slider and hover to view the specific values for each day. In addition, the visualization interface also includes an equipment management view (e.g., ... Figure 5 , Figure 6 As shown in the figure, this view displays the device number, device type, maintenance manager, and current working status of all registered devices in a standardized table. It supports fuzzy search by device number or name, filtering by type, and adding or deregistering devices. The device status indicator is linked to the monitoring device's heartbeat packet in real time. When a device goes offline or reports a fault code, the corresponding row status cell refreshes the color block and text.
[0047] The early warning module, as a backend microservice of the platform, shares a real-time indicator library with the data processing module and provides a front-end rule configuration interface (such as...). Figure 7b (As shown). Managers can use this interface to set threshold-based early warning rules for different equipment operation indicators, such as "failure risk index higher than 80% for 5 minutes" or "real-time fuel consumption exceeds the benchmark value by 30%". Rules can be refined to specific equipment, equipment type, or construction procedure. The early warning module's background daemon periodically polls the real-time indicator stream, comparing each new data point with all activated rules. Once the trigger condition is met, a structured early warning message is immediately generated, including the triggering equipment identifier, rule name, actual value, threshold, and timestamp. This early warning message is pushed to the early warning center dashboard on the visual interface (e.g., ...). Figure 7a As shown in the figure, it is highlighted in a list and automatically classified and statistically analyzed according to warning type, associated equipment, and associated process; on the other hand, it is forwarded to the scheduling module in real time via WebSocket.
[0048] The dispatch module is integrated into the platform's command center interface, and its core function is a one-click real-time intercom communication function. When the early warning module issues an early warning, the dispatch module highlights the device icon that triggered the warning with a flashing red circle on the map module of the visualization interface, and displays an "Immediate Intercom" button in the device details card. Managers can click this button to establish a full-duplex voice link with the vehicle-mounted terminal of the corresponding construction equipment in the tunnel through the platform's built-in real-time intercom communication module. During the call, managers can refer to the trends in equipment operation indicators displayed in the chart module and the distribution of surrounding equipment presented in the map module to issue specific dispatch instructions to operators, such as "Stop tunneling, check for abnormal fuel consumption" and "Transfer to support the No. 2 working face." The dispatch module automatically records the early warning response time, intercom initiation time, and instruction content for each warning, forming a complete management loop from anomaly detection and early warning triggering to manual intervention. All records can be replayed along a timeline in the post-event traceability interface of the data dashboard.
[0049] Through the collaborative work of the five modules mentioned above—monitoring, processing, early warning, visualization, and scheduling—the entire smart construction site big data platform achieves full-chain digital management of construction equipment operation data in the enclosed environment of a tunnel, significantly improving the real-time performance and accuracy of equipment utilization assessment, cost accounting, risk prediction, and emergency dispatch.
[0050] In an embodiment of the present invention, the early warning module includes a rule configuration interface and an early warning statistics dashboard; the rule configuration interface is used for users to customize and set threshold conditions for different equipment operation indicators to form early warning rules; the early warning statistics dashboard is used to classify, statistically analyze and display the triggered early warning information according to the early warning type, associated equipment and associated process.
[0051] In its implementation, the early warning module is presented as a sub-function of the smart construction site big data platform. Its core consists of two parts: a rule configuration interface and an early warning statistics dashboard, both of which are integrated into the platform's visualization interface system.
[0052] The rule configuration interface is embedded in the platform's function navigation area as a form-style page. Users can access this interface under the "Early Warning Center" menu after logging into the platform with an account that has administrator privileges. The top of the interface features a rule list area, which displays all currently effective early warning rules in a table. It includes fields such as rule name, applicable device scope, monitoring indicators, threshold conditions, creator, and last modification time, and supports filtering by rule status (enable / disable). The middle of the interface is the rule editing area, where users can complete the full definition of early warning rules. First, users select the equipment operation indicators to be monitored from the "Monitoring Indicators" drop-down box. The options in this drop-down box are synchronized with the data processing module in real time. Preset indicators include equipment utilization rate index, cost control index, failure risk index, and raw data collection indicators such as real-time fuel consumption, continuous idling time, and engine water temperature. After selecting an indicator, the interface dynamically expands the threshold setting component. This component provides a drop-down list of comparison operators (greater than, less than, equal to, within the interval, outside the interval, etc.) and a numerical input box. Some indicators support both percentage threshold and absolute value threshold modes. Subsequently, users limit the equipment objects to which the rules apply through the "Effective Scope" selector. Three granularities are supported: single equipment (searched by equipment number), equipment type (such as loader, transport engineering vehicle, wet spraying trolley), or construction process (such as slag removal, support, lining). Multiple selections can be combined to achieve composite conditions.
[0053] In addition, the rule configuration interface provides a "Duration" setting, where users can enter a value in seconds. An alert will only be triggered after the indicator continuously meets the threshold conditions for this duration, effectively avoiding false alarms caused by instantaneous fluctuations. After all settings are filled in, the user clicks the "Save Rule" button. The front-end encapsulates the above configuration parameters into a JSON-formatted rule object and sends it to the back-end microservice of the alert module via Hypertext Transfer Protocol. The back-end service parses the rule object and writes it into the alert rule table of the relational database, while simultaneously loading the rule into the rule engine instance in memory. At this point, the rule enters real-time monitoring mode.
[0054] The early warning statistics dashboard, as a separate page, shares the same database as the rule configuration interface, presenting a real-time statistical view of all early warning events. The dashboard uses a responsive layout. The top section features key indicator cards, dynamically displaying the total number of new early warnings today, the number of unprocessed early warnings, the average response time, and the number of devices currently in a high-risk state. The left-hand side of the middle section is a categorized statistical visualization area, using stacked bar charts to show the hourly distribution frequency of various early warning types (such as abnormal fuel consumption, idle timeout, fault risk, and location violation) over the past 24 hours, a pie chart to show the percentage of early warnings triggered by different device types, and a horizontal bar chart to show the top five construction procedures with the highest early warning frequency. The right-hand side of the middle section is an early warning event stream list, using an infinite scrolling loading method to display early warning records one by one in reverse chronological order of trigger time. Each record includes the early warning type, device number, device type, current value, threshold, trigger time, and a "Confirm" button. Unconfirmed early warnings are highlighted with a yellow background, while confirmed early warnings are displayed on a white background with the confirmer and confirmation time indicated. The dashboard features a control area at the bottom, offering multi-dimensional filtering tools: users can combine filtering by alert type (all, fuel consumption, idling, fault, out of bounds), associated equipment (enter equipment number for fuzzy matching), and associated process (select process dictionary item) via drop-down menus. The filtered results are updated in real-time with a list and visual charts. To the right of the filter is an "Export Report" button, supporting the export of alert statistics results under the current filtering conditions to Excel or CSV files. Fields include alert number, triggering equipment, alert type, threshold, actual value, first trigger time, last trigger time, duration, associated process, and processing status. The backend implementation of the alert statistics dashboard relies on an alert history table maintained by the alert module. This table uses the unique alert number as the primary key, generating a record each time the rule engine triggers an alert, and continuously updating the last trigger time and duration of the same persistent event. The backend provides aggregated query services via a RESTful interface, while the frontend asynchronously polls or receives real-time push notifications via WebSocket to ensure that dashboard data is updated synchronously with alert events.
[0055] Through the collaborative work of the aforementioned rule configuration interface and early warning statistics dashboard, the early warning module achieves full-process functional support for quantifying manual management experience into executable logical rules and for automatically monitoring and multi-dimensional review and analysis of abnormal states of tunnel construction equipment.
[0056] In embodiments of the present invention, the visualization interface further includes an equipment management view, which is used to display the equipment number, equipment type, operation and maintenance person in charge and current working status of all registered construction equipment in a list form, and supports managers to query, filter, add and delete construction equipment by equipment number or equipment name.
[0057] The smart construction site big data platform's visual interface includes an embedded equipment management view. This view presents the digital archives of all registered construction equipment within the tunnel in a standardized data table format, serving as the unified entry point for the platform's equipment asset management. The equipment management view is located under the "Equipment Center" primary menu in the platform's navigation bar. The page uses a two-column layout. The left side displays an equipment type tree, hierarchically categorizing and collapsing the quantity of various equipment types, such as tunnel boring machines, loaders, transport vehicles, and wet spraying trolleys. The right side is the main data table area, with column headers displaying four core fields from left to right: equipment number, equipment type, maintenance manager, and current working status. Among them, the equipment number is a unique asset identifier assigned by the system to each construction equipment, supporting numeric, alphanumeric, and combined encoding formats. It is automatically generated by the data processing module and written to the equipment master table when the equipment is first connected to the platform. The equipment type is taken from the platform's pre-set dictionary table, which can be extended by users through backend configuration. The maintenance manager field stores the name and employee number of the current responsible person. This relationship is maintained through an organizational structure table synchronized via WeChat or a lightweight directory access protocol, supporting multiple people to take turns on a single equipment. The current working status field is linked in real time with the heartbeat packets and status data frames of the monitoring device deployed on the equipment. The monitoring device reports the heartbeat every thirty seconds. The data processing module comprehensively determines the equipment's status as one of four states: on-machine, offline, standby, or faulty, based on parameters such as the latest heartbeat reception time, engine start / stop status, fault code register value, and actuator current, and writes the determination result back to the corresponding field in the equipment master table.
[0058] A search bar and function button area are deployed above the table. To the left of the search bar are input boxes for device number and device name, which are logically linked by an "OR" relationship. After an administrator enters a character in either input box, the front end triggers a debounced delayed query, sending a structured query request containing the keyword to the backend data interface. The backend performs a joint search in the main device table, performing exact matching of the device number and fuzzy matching of the device name. The search results are returned as a JSON data packet. The front end uses virtual scrolling technology to redraw the table data without refreshing the page, with the entire process response time controlled within 300 milliseconds. To the right of the search bar is a filter dropdown menu. Menu options are dynamically bound to the device type dictionary and the maintenance manager list. When a user selects conditions such as "Show only loaders" or "Show only devices managed by Zhang San," the front end immediately appends the filter parameters to the data request, and the table only renders data rows that meet the conditions. The function button area includes two frequently used operation entry points: "Add Device" and "Batch Delete." When a user clicks the "Add Device" button, a modal form window pops up. The form includes input fields such as device number, device type, operations and maintenance manager, initial working status (default is "offline"), and remarks. The device number is validated in real-time to ensure it is not duplicated with existing records. The device type is presented as a drop-down menu, and the operations and maintenance manager is asynchronously searched from the organizational structure component. After the user completes and submits the form, the front end performs mandatory field validation and format verification. Upon successful validation, a Hypertext Transfer Protocol (HTTP) request is sent to the data processing module. The data receiving end first writes the new device information into the device master table and simultaneously triggers the monitoring device registration process: the platform generates a unique device key based on the device number and sends an activation command to the corresponding monitoring device via the MQTT protocol. After receiving the command, the monitoring device reports its first heartbeat, completing the binding between the physical device and the digital file. The new device record is then inserted into the first row of the table in real-time, and the count statistics of the device type tree on the left are updated synchronously.
[0059] The deletion operation supports both single-record deletion and batch deletion. Each data row ends with a red "Delete" icon button. Clicking this button triggers a secondary confirmation bubble prompting, "Deleting the device will simultaneously unbind the monitoring device and clear historical warning rules. Continue?" Batch deletion involves selecting multiple records using the checkbox in the first column of the table, which then activates the "Delete" button in the top toolbar. After user confirmation, the front-end encapsulates the list of device numbers to be deleted and sends it to the back-end. The back-end performs cascading operations within a database transaction: First, it sets the "Cancellation Flag" field of the corresponding record in the device's master table from 0 to 1, achieving logical deletion while retaining all historical operating data, trajectory data, and warning records for audit traceability. Second, it sends an unbinding command to the monitoring device, including a retention period parameter. Upon receiving the command, the monitoring device stops heartbeat reporting and enters a pending configuration state, but locally cached historical data remains readable until the next activation. Finally, it clears all warning rules associated with the device in the rule engine and the corresponding entries in the rule condition mapping table, ensuring that the cancelled device no longer triggers any warnings. After a successful deletion, the front end receives the device change notification broadcast by the server via WebSocket, automatically removes the corresponding data row from the table, and decrements the quantity value in the device type tree. The entire process does not require the user to manually refresh the page.
[0060] The equipment management view maintains strong real-time synchronization with other modules of the platform. When a monitoring device experiences a heartbeat interruption for more than 90 seconds due to tunnel signal obstruction, the data processing module determines that the device is offline, immediately updates the current working status field of the device master table to "offline," and records a timestamp. Simultaneously, it pushes a status change message to all online front-ends via WebSocket. Upon receiving the message, the equipment management view locates the corresponding device row, changes the text and left-side color block in the "Current Working Status" cell of that row from green "Online" to gray "Offline," and automatically adds an "Offline Timeout" marker to the remarks column. This status synchronization mechanism ensures that management personnel, whether located on the command center's large screen or a mobile device, can accurately and in real-time grasp the registered list, assigned personnel, and operational health status of all construction equipment within the tunnel through the equipment management view, providing a single, reliable master data source for subsequent scheduling decisions and preventative maintenance.
[0061] In an embodiment of the present invention, the device management view further includes a device status identifier that is linked in real time with the monitoring device. The device status identifier is used to distinguish and display the current working status as online, offline, standby, or fault on the visualization interface.
[0062] The smart construction site big data platform's equipment management view incorporates a real-time equipment status identification mechanism linked to monitoring devices. This mechanism, combining graphical and textual methods, clearly distinguishes the current working status of each construction device on the visual interface, indicating whether it is online, offline, standby, or faulty. The equipment management view uses a standardized table layout, with each row corresponding to one registered construction device. The table column headers consistently display the device number, device type, maintenance manager, and current working status. The current working status column is the core area for displaying equipment status identification. Each cell in this column contains a combination of a circular color block icon on the left and a status name text on the right. The circular color block is drawn using scalable vector graphics, with a diameter of twelve pixels. Online status is filled with green and accompanied by the text "Online," offline status with gray and accompanied by the text "Offline," standby status with yellow and accompanied by the text "Standby," and faulty status with red and accompanied by the text "Fault." A four-pixel spacing is maintained between the color block and the text to ensure clear visual recognition on both the command center's large screen and mobile terminals.
[0063] The real-time linkage logic of the equipment status identification is completed collaboratively by the data processing module and the front-end visualization engine. The monitoring devices deployed on each construction equipment in the tunnel send heartbeat packets to the platform at fixed intervals (30 seconds by default). The heartbeat packets contain at least the equipment's unique identifier, timestamp, and the equipment's main controller self-test status flag. At the same time, the monitoring devices continuously collect operating parameters from the equipment controller's local area network bus, including engine speed, hydraulic pump current, fault code register values, and infrared sensor signals from the control room, and encapsulate these parameters into status data frames for asynchronous uploading. Upon receiving a heartbeat packet or status data frame, the data processing module immediately triggers a status determination process: if the current system time minus the time of receiving the latest heartbeat packet from the device exceeds ninety seconds, it is determined to be in an offline state; if the heartbeat is normal and the engine speed is greater than zero, but the current of all actuators remains below the no-load threshold for more than five minutes, and the infrared sensor in the control room detects no one for thirty consecutive minutes, it is determined to be in a standby state; if the heartbeat is normal and the device reports any serious fault code through the controller local area network bus, or the temperature sensor and vibration sensor values built into the monitoring device exceed the preset hardware protection threshold, it is determined to be in a fault state; if the heartbeat is normal and none of the above offline, standby, or fault conditions are met, it is determined to be in an online state. After each status determination, the data processing module immediately writes the determination result into the current working status field of the device's main table and appends a status change timestamp.
[0064] Meanwhile, the data processing module pushes device status change messages to all online front-ends via the WebSocket protocol. The message body uses a lightweight JSON format, including the device number, the updated status value, and the status change time. The front-end of the device management view subscribes to this WebSocket channel. When receiving a status change message, it first locates the corresponding row index in the table dataset according to the device number in the message. Subsequently, through the virtual DOM differential update algorithm, it only replaces the color block filling color and status text in the "current working status" cell of this row, without re-requesting the entire table data or triggering the redrawing of other irrelevant rows, thus achieving flicker-free status synchronization. For the scenario of first loading the device management view or when the user manually refreshes the page, the front-end initiates a RESTful API request to the back-end to obtain the full device list and the latest status snapshots of each device. The dataset returned by the back-end already contains the real-time current working status field in the device main table. The front-end completes the initial rendering based on this and dynamically generates corresponding color block and text combinations for each row during the rendering process.
[0065] This device status identification mechanism also supports multi-dimensional status filtering and statistics. In the filtering drop-down menu at the top of the device management view, there is a "by working status" filtering option built in. After the user checks "online", "offline", "standby", or "fault", the front-end immediately attaches status filtering parameters to the data interface. The back-end executes an exact match of the status field in the structured query statement and only returns the data rows that meet the conditions. The status identification color blocks and text in the filtering results still maintain real-time linkage capabilities. In addition, the data dashboard module of the platform aggregates the status statistics of all devices and displays the quantity and proportion of the four statuses of online, offline, standby, and fault among the devices registered at the current moment in a circular chart. The data source of this circular chart is homologous to the status determination result of the device management view and both come from the current working status field of the device main table maintained by the data processing module, ensuring the consistency of the status semantics across the entire platform.
[0066] Through the above real-time linkage design of device status identification, managers do not need to enter the detail page of each device. By simply observing the color of the color blocks in each row of the device management view table, they can master the online situation and operation health of all construction devices in the tunnel. The four-color system of green, gray, yellow, and red directly corresponds to the availability level of the devices, providing an immediate and accurate decision-making basis for subsequent early warning handling, dispatching assignment, and maintenance scheduling.
[0067] In an embodiment of the present invention, the scheduling module is further configured to generate the historical movement trajectory of construction equipment based on location data and display the historical movement trajectory on the map module.
[0068] The scheduling module of the intelligent construction site big data platform is deeply integrated into the map module and the device data interface of the visualization interface (such as Figure 8As shown, the monitoring device (as shown) is responsible for generating and visualizing the movement trajectory based on the historical location data of the equipment. The monitoring devices deployed on various construction equipment within the tunnel continuously collect location data. Each frame of location data includes the equipment number, longitude coordinates, latitude coordinates, elevation, instantaneous speed, and collection timestamp, and is transmitted back to the data processing module in real time via an anti-interference data transmission module. The data processing module persistently stores all location data in a distributed database that supports spatiotemporal queries, indexed by both equipment identifier and timestamp, forming a complete historical archive of equipment movement.
[0069] When administrators need to perform retrospective analysis on the historical operation paths of a particular device, they can perform the following operations on the platform interface: Enter the device data interface (e.g., Figure 8 As shown in the image, the left side of the interface displays a device list tree, while the right side contains a map module and a trajectory control panel. Administrators first search for and select the target construction equipment in the device list by device number or name. Then, on the trajectory control panel, they use the date picker to set the start and end times for the query, supporting filtering by hour, day, or custom time period. Clicking the "Load Trajectory" button encapsulates the device's unique identifier and time window parameters into an HTTP request and sends it to the trajectory service interface on the scheduling module's backend.
[0070] Upon receiving the request, the scheduling module immediately performs a structured query in the spatiotemporal database, retrieving all location records with acceptable positioning accuracy for the device within the specified time period. Abnormal points with a positioning factor greater than five or an inertial navigation cumulative error exceeding one meter when in a base station handover blind zone are removed. The search results are arranged in ascending order of time, forming the original trajectory point sequence. To optimize the front-end rendering effect and trajectory smoothness, the scheduling module performs uniform interpolation to fill in point pairs with adjacent time intervals greater than thirty seconds. The interpolation position is calculated linearly based on the latitude, longitude, and time difference of the preceding and following points, ensuring that the trajectory line is continuous and uninterrupted on the time axis. The processed trajectory point sequence is encapsulated in GeoJSON format and returned to the front-end map module as the response payload.
[0071] The map module, based on a high-precision electronic map of the tunnel, sequentially connects the received trajectory point sequences to generate a polyline vector layer. The width of the trajectory polyline is set to three pixels, and the color uses a gradient mapping from cool to warm colors, dynamically assigned based on the calculated movement speed between adjacent points: low-speed range (0~3km / h) is displayed in blue, medium-speed range (3~8km / h) in orange, and high-speed range (above 8km / h) in red. A green circular marker is overlaid on the trajectory start point and automatically labeled "Start," while a red square marker is overlaid on the end point and labeled "End." When the user hovers the mouse over any position on the trajectory line, the map module pops up an information pop-up window, displaying the absolute acquisition time, relative time, instantaneous speed, and distance from the tunnel entrance in real time. The trajectory control panel synchronously provides a player control, supporting playback of trajectory animations at speeds from five to sixty times. During playback, the trajectory segment dynamically extends based on the current timestamp, and the device's current position icon moves synchronously along the trajectory line; traveled segments are displayed as solid lines, and untraveled segments are hidden.
[0072] While generating the trajectory line, the scheduling module performs real-time post-processing on the trajectory data to extract high-value semantic information, which is then automatically displayed as text labels around the trajectory line. The dwell point identification algorithm is as follows: if the equipment's position changes by less than two meters within ten consecutive minutes and the engine is running, it is determined to be a valid work dwell. The system records the start time, end time, and center coordinates of the dwell, and anchors it near the corresponding position on the trajectory line with a bubble label "Dwelling for XX minutes". The work range analysis is generated based on the minimum bounding rectangle or convex hull polygon of the trajectory point set, and rendered on the map as a semi-transparent color block to intuitively show the boundary of the equipment's activity area within the time period. The work cycle count is automatically counted based on the closed-loop characteristics of the trajectory formed by the equipment repeatedly traveling back and forth between the tunneling face and the slag outlet. Each time from point A to point B and back to point A is counted as a complete cycle, and the number of cycles is displayed in the trajectory information summary column.
[0073] The aforementioned historical movement trajectories and their derived analysis results not only support static viewing on the current interface but also support exporting to common vector format files (such as KML and GeoJSON) or static images, facilitating their inclusion in construction logs, progress reports, and efficiency evaluation meetings. Furthermore, the scheduling module associates a real-time intercom entry with each trajectory point and stop point on the trajectory display interface: managers can right-click on any trajectory point or stop label on the map to bring up a "Talk to this device" button, initiating real-time voice communication with the corresponding construction equipment operator to inquire about the current working conditions or issue subsequent scheduling instructions. Through this mechanism, the scheduling module reconstructs discrete location data into a traceable, quantifiable, and interactive panoramic view of movement paths, enabling managers to clearly understand the compliance of the operating paths of each construction device within the tunnel, the proportion of ineffective movements, work efficiency levels, and space resource occupancy.
[0074] In embodiments of the present invention, the data processing module is also used to perform time-series analysis on the historical operating status data and historical fuel consumption data of the construction equipment, generate equipment failure probability prediction and remaining service life prediction, and output the prediction results through a visual interface.
[0075] The data processing module of the smart construction site big data platform embeds a predictive analytics engine. This engine continuously reads historical operating status data and historical fuel consumption data of construction equipment stored in a time-series database. Based on time-series analysis and machine learning models, it generates a failure probability prediction and a remaining service life prediction for each piece of equipment, and displays the data analysis through a visual interface dashboard (such as...). Figure 9 (As shown) Output the prediction results. The data processing module first filters out the active device identifiers from the device master table. For each device, it extracts the operating status data and corresponding fuel consumption data for at least twelve months from the time series database, indexed by the device number and ordered by the collection timestamp. The operating status data includes at least engine speed, hydraulic oil temperature, vibration acceleration amplitude, frequency of controller local area network bus fault codes, real-time working status codes, and cumulative working time; the fuel consumption data covers instantaneous fuel consumption and cumulative fuel consumption per unit time. After the raw data stream enters the preprocessing pipeline, outlier removal based on the interquartile range method is first performed to remove outliers exceeding three times the interquartile range. Cubic spline interpolation is used to fill in the missing timestamp segments caused by communication interruptions. Finally, the entire data is resampled into an equally spaced time series of one per minute. Subsequently, the predictive analysis engine segments the continuous series with a fixed sliding time window. The window length is preset to seven days, and the sliding step is twenty-four hours. Each window generates a multi-dimensional feature vector. The feature extraction module calculates the time-domain statistics of each monitoring indicator within each time window, including mean, standard deviation, maximum, minimum, peak-to-peak value, root mean square, and first-order difference mean. For high-frequency signals such as vibration acceleration, the module extracts the first five spectral energies and the amplitude of the main frequency band as frequency domain features using Fast Fourier Transform. Simultaneously, it extracts economic features such as unit fuel consumption output ratio and fuel consumption fluctuation coefficient from the fuel consumption sequence. These feature vectors, together with the corresponding equipment labels, constitute the training sample set.
[0076] Fault probability prediction employs a gradient boosting decision tree algorithm to construct a binary classification model, using "whether a serious fault leading to downtime will occur within the next seven days" as the prediction label. Fault determination is based on the equipment reporting a serious downtime fault code or continuous downtime exceeding four hours after engine shutdown. During model training, five-fold cross-validation and an early shutdown strategy are used to prevent overfitting. Feature importance ranking is calculated using the average reduction in Gini impurity, outputting the contribution weight of each feature to the prediction result. After model training, the model is persisted to a model repository. In the real-time prediction phase, the data processing module populates the latest time window with the equipment's operating status data and fuel consumption data from the past seven days. After the same preprocessing and feature extraction process, a real-time feature vector is generated and input into the trained fault probability model, outputting a fault probability value between 0% and 100%. This probability value is updated every 24 hours and concatenated with historical probability values to form a time-series prediction curve.
[0077] The remaining useful life prediction employs a similarity-based degradation trajectory matching method. The system maintains a reference sample library, storing daily feature vector sequences of historically failed devices for the 365 days prior to their first failure. Each sequence indicates the actual number of days remaining for the device from that day until the first failure. For the device to be predicted, its feature vector sequence from the past 30 days is extracted as the query sequence. A dynamic time warping algorithm is used to calculate the similarity distance between this query sequence and each historical degradation trajectory in the reference sample library; a smaller distance indicates a closer similarity in degradation patterns. The system selects the ten most similar historical trajectories, using the reciprocal of the similarity distance as weight, and calculates the weighted average of the remaining useful life corresponding to these ten trajectories. This average is used as the predicted remaining useful life value for the current device, accurate to the hour. Each prediction also outputs a 90% confidence interval, the width of which is determined by the lifespan dispersion of the selected similar trajectories. To maintain model timeliness, the system performs incremental training on the full sample set monthly, dynamically updating the feature weights and the reference sample library.
[0078] The prediction results are presented through a data analysis dashboard with a visual interface (such as...). Figure 9(As shown) This dashboard outputs information to management personnel. It uses a card-based layout to display predicted information by device. Each card displays the device number and type at the top. A semi-circular dashboard on the left side of the center indicates the current failure probability in real time. The dial pointer angle is linearly mapped to the probability value. The dashboard background color gradually changes from green (0%) to red (100%), and is marked with three threshold segments: "Low Risk," "Medium Risk," and "High Risk." The default thresholds are 30% and 70%, which can be manually adjusted. A horizontal progress bar on the right side of the center displays the remaining service life. The length of the progress bar represents the proportion of the predicted remaining time to the design baseline life of the same type of equipment. The baseline life is taken from the rated life on the equipment nameplate or industry standards. The end of the progress bar displays the remaining hours and the expected failure date in numerical form. The bottom of the card features a collapsible "Trend Details" panel, which, when unfolded, displays a line graph of the failure probability change over the past 30 days, a curve showing the decline in the remaining life prediction trend, and a bar chart ranking the contribution of each feature parameter to the current prediction result. The contribution ranking is displayed side-by-side with the feature name and percentage value, allowing managers to identify key factors affecting equipment health, such as "excessive vibration amplitude" or "increased fuel consumption fluctuations." The top of the dashboard provides a global filter bar, supporting combined filtering by equipment type, failure risk level (low / medium / high), and remaining life threshold range (e.g., less than 500 hours), quickly locating target equipment requiring priority preventative maintenance. All prediction results and feature data can be exported to PDF or Excel format for use in maintenance planning meetings and spare parts procurement decisions. Through this mechanism, the data processing module transforms massive amounts of historical operating data and fuel consumption data into quantitative estimates of future failure risks, shifting tunnel construction equipment management from passive, reactive maintenance to proactive, condition-based maintenance, significantly reducing unplanned downtime probability and total lifecycle maintenance costs.
[0079] In an embodiment of the present invention, the early warning rules received by the early warning module for equipment operation indicators include: differentiated thresholds set for different construction procedures, differentiated thresholds set for different geological conditions, and differentiated thresholds set for different equipment types.
[0080] The early warning module of the smart construction site big data platform supports setting differentiated early warning thresholds for different construction procedures, geological conditions, and equipment types. This allows the early warning rules to accurately adapt to the complex working environment of tunnel excavation, characterized by dynamic changes in working conditions, alternating surrounding rock grades, and varying equipment performance. This differentiated threshold configuration function is integrated into the rule configuration interface of the early warning module (e.g., ...). Figure 7bAs shown in the image, the interface adds an "Effective Conditions" configuration area in addition to the basic early warning rule form. This area contains three independent drop-down selectors, which are respectively associated with the platform's pre-set construction procedure dictionary, geological condition dictionary, and equipment type dictionary. The construction procedure dictionary is pre-imported based on the tunnel construction organization design documents and covers the entire process chain, including face drilling and blasting, muck removal and transportation, initial support, invert arch pouring, waterproof layer laying, and secondary lining. Each procedure corresponds to a unique procedure identifier and mileage station range. The geological condition dictionary is established based on geological exploration data and is divided into five categories according to the surrounding rock grade: Grade I, Grade II, Grade III, Grade IV, and Grade V. It can be extended to special geological labels such as fault fracture zones, water-rich layers, gas areas, and soft rock large deformation sections. Each geological segment is bound to the tunnel mileage linear reference system. The equipment type dictionary is from the same source as the equipment type field in the equipment management view, including tunnel boring machines, loaders, transport engineering vehicles, wet spraying trolleys, grouting machines, and anchor drilling rigs, and supports further subdivision by equipment model.
[0081] When creating or editing early warning rules, managers first select the equipment operation indicators to be monitored from the monitoring indicator drop-down box, such as "fuel consumption per unit time," "engine coolant temperature," "failure risk index," or "continuous idling duration." Then, they fill in the basic threshold value and comparison operator in the threshold setting component. Next, they check the "Enable differentiated thresholds" checkbox in the effective condition configuration area and combine and set the effective conditions according to management needs. For example, managers can create three differentiated rules for the "fuel consumption per unit time" indicator: the first rule limits the effective process to "slag transportation," with a threshold upper limit set at 18 liters / hour; the second rule limits the effective geology to "Class V surrounding rock," with a threshold upper limit set at 25 liters / hour; and the third rule limits the effective equipment type to "loader," with a threshold upper limit set at 15 liters / hour. Each differentiated rule can independently set its duration, early warning level, and notification recipients, and supports quickly generating similar configurations via the "copy rule" button. Once all settings are complete, the front-end encapsulates the rule object into a JSON structure containing indicator identifiers, threshold values, a list of effective procedures, a list of effective geological conditions, a list of effective equipment types, and priority weights, and sends it to the back-end of the early warning module via the Hypertext Transfer Protocol.
[0082] The early warning module maintains a rule condition mapping table in the backend. The table structure uses the rule identifier as the primary key and stores monitoring indicators, comparison operators, threshold values, effective process identifiers, effective geological identifiers, effective equipment type identifiers, and priority weight fields. When the data processing module pushes real-time equipment operation indicator streams, the early warning module first extracts the equipment identifier and the current indicator value from the data frame. Then, it uses the equipment identifier as an index to concurrently query the real-time operating context of the equipment. The determination of the current construction process depends on the spatial inclusion relationship between the equipment's real-time location coordinates and the polygons representing the process segment in the tunnel electronic map: the map module maintains a closed polygon or linear reference interval corresponding to each process. The early warning module inputs the equipment's current coordinates through a spatial index interface and returns the process identifier to which those coordinates belong. The determination of the current geological conditions uses a similar mechanism: the system pre-marks the geological types of each segment along the tunnel mileage markers. The early warning module matches the corresponding mileage markers based on the equipment's location coordinates to obtain the geological segment identifier to which that marker belongs. The equipment type is an inherent attribute statically stored in the equipment master table and is directly read from the equipment file.
[0083] After obtaining the three types of context tags mentioned above, the early warning module uses them as query conditions and performs multi-condition matching retrieval in the rule condition mapping table. The retrieval logic is as follows: all rules whose effective process identifier matches the current process of the equipment, or whose effective geological identifier matches the current geological condition of the equipment, or whose effective equipment type identifier matches the equipment type are selected, including global rules without any specified effective conditions. If the same equipment operation indicator hits multiple differentiated rules at the same time, the early warning module will make conflict resolution based on the priority weight field. The priority weight is manually assigned by the administrator during configuration, with a value range of 1 to 100, and the larger the value, the higher the priority. If the weights are the same, a unique effective rule is selected according to the default arbitration order of "process takes precedence over geology, geology takes precedence over equipment type, and equipment type takes precedence over global rules". After the rule is selected, the early warning module compares the real-time indicator value with the threshold of the rule. If the threshold condition is met and the duration reaches the target, an early warning message is generated, and the effective condition type and specific tag value on which this early warning is based are explicitly marked in the data structure of the early warning message, such as "effective condition: process: slag transportation" or "effective condition: surrounding rock: grade V".
[0084] Early warning statistics dashboard (such as) Figure 7a(As shown) When displaying such differentiated warnings, a "Effective Conditions" column is added to the warning list to clearly present the triggering context of each warning in the form of labels. The category statistics area at the top of the dashboard supports drill-down analysis by the dimension of effective conditions: when the user selects "Statistics by Process", the system aggregates and displays the number and percentage of warnings triggered by each construction process; when the user selects "Statistics by Surrounding Rock Grade", a heat map of warning distribution under different geological conditions is displayed; when the user selects "Statistics by Equipment Type", the warning frequency of various types of equipment is displayed in order. All statistical charts support linked filtering. Clicking on a bar chart block of a specific process will immediately filter the warning list to show warning records triggered only under that process.
[0085] Through the aforementioned differentiated threshold configuration and context-aware early warning mechanism, the early warning module effectively avoids the problem of missed or false alarms caused by fixed thresholds due to frequent alternation of processes, drastic geological changes, and differences in equipment performance during tunnel construction. This ensures that equipment anomaly identification truly matches the real-time working conditions at the tunnel face, significantly improving the accuracy and usability of the early warning.
[0086] like Figure 1 As shown, this embodiment of the invention also provides an operation and management method for tunnel construction equipment, including: The monitoring device collects real-time operating status data, fuel consumption data, and location data of the construction equipment, and uses an anti-interference data transmission module to transmit the operating status data, fuel consumption data, and location data to the data processing module. The data processing module integrates and analyzes the received operating status data, fuel consumption data, and location data to generate equipment operation indicators. The system acquires equipment operation indicators in real time through the early warning module, compares these indicators with user-defined early warning rules, and automatically generates and issues early warning information when the early warning conditions are met. The real-time location of construction equipment and the changing trends of equipment operation indicators are displayed through a visual interface. The scheduling module receives early warning information from the early warning module, and based on the changing trends of operational indicators, the management personnel conduct voice communication with the operators of construction equipment in the tunnel to execute scheduling instructions.
[0087] The above provides a detailed description of the operation management system and method for tunnel construction equipment. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.
Claims
1. An operation and management system for tunnel construction equipment, characterized in that, The system includes: The monitoring device is deployed on the construction equipment inside the tunnel to collect real-time operating status data, fuel consumption data, and location data of the construction equipment. The operating status data, fuel consumption data, and location data are transmitted to the data processing module using an anti-interference data transmission module. The anti-interference data transmission module includes an inertial measurement unit, an ad hoc network communication unit, a hybrid backhaul unit, a tunnel signal dynamic sensing engine, and a multi-link intelligent collaborative controller. The inertial measurement unit maintains the continuity of position data through dead reckoning when satellite positioning signals fail. The ad hoc network communication unit establishes an ad hoc network with adjacent monitoring devices and transmits data via multi-hop relays. The hybrid backhaul unit forms a hybrid backhaul link with a wireless network via a leaky coaxial cable. The tunnel signal dynamic sensing engine collects multi-link signal quality parameters of each monitoring device's location in real time, constructing a dynamic signal heatmap of the entire tunnel. The multi-link intelligent collaborative controller allocates transmission links for data of different priorities based on the dynamic signal heatmap and optimizes the ad hoc network topology in real time. The data processing module is used to integrate and analyze the received operating status data, fuel consumption data, and location data to generate equipment operation indicators; The early warning module is used to compare the real-time acquired equipment operation indicators with the early warning rules set by the user, and automatically generate and issue early warning information when the early warning conditions are met. The visualization interface includes a map module for displaying the real-time location of the construction equipment and a chart module for displaying the changing trends of equipment operation indicators; The scheduling module is used to receive the early warning information and support managers to conduct voice communication with the operators of construction equipment in the tunnel to execute scheduling instructions based on the changing trends of the operational indicators.
2. The system according to claim 1, characterized in that, The operating status data includes at least one of the following: average daily working time of the equipment, real-time operating status, operating rate, and working efficiency; the fuel consumption data includes at least one of the following: average daily fuel consumption of the equipment and real-time fuel consumption; the location data includes at least one of the following: real-time location coordinates of the equipment and historical movement trajectory.
3. The system according to claim 2, characterized in that, The equipment operation indicators generated by the data processing module include at least one of the following: equipment utilization rate index, cost control index, and failure risk index.
4. The system according to claim 3, characterized in that, The early warning module includes a rule configuration interface and an early warning statistics dashboard. The rule configuration interface is used for users to customize threshold conditions for different equipment operation indicators to form early warning rules. The early warning statistics dashboard is used to classify and display the triggered early warning information according to the early warning type, associated equipment, and associated processes.
5. The system according to claim 4, characterized in that, The visualization interface also includes an equipment management view, which displays the equipment number, equipment type, maintenance manager, and current working status of all registered construction equipment in a list format. It also supports managers to query, filter, add, and delete construction equipment by equipment number or equipment name.
6. The system according to claim 5, characterized in that, The device management view also includes a device status identifier that is linked in real time with the monitoring device. The device status identifier is used to distinguish and display the current working status as online, offline, standby, or fault on the visualization interface.
7. The system according to any one of claims 1 to 6, characterized in that, The tunnel signal dynamic perception engine has a built-in signal quality acquisition unit. The signal quality acquisition unit is used to continuously collect satellite signal strength, self-organizing network link signal-to-noise ratio, and leakage coaxial signal attenuation value, generate a three-dimensional signal fingerprint containing location, time, and signal quality, and upload it to the edge computing node in the tunnel. The edge computing node constructs and updates the dynamic signal heat map of the entire tunnel according to the signal fingerprint at a preset frequency.
8. The system according to claim 7, characterized in that, The multi-link intelligent collaborative controller divides the data to be transmitted into four priorities: control commands, early warning information, status data, and historical logs; and divides the data into signal-good areas, signal-weak areas, and signal-blind areas according to the dynamic signal heat map, so as to match the corresponding transmission links for data of different priorities.
9. The system according to claim 8, characterized in that, The multi-link intelligent collaborative controller is also used to calculate the adaptive topology of the self-organizing network in real time based on the real-time movement trajectory of the construction equipment and the dynamic signal heat map, and to automatically adjust the transmission power and relay path of the self-organizing network communication unit.
10. A method for the operation and management of tunnel construction equipment, characterized in that, The method includes: The monitoring device collects real-time operating status data, fuel consumption data, and location data of the construction equipment, and uses an anti-interference data transmission module to transmit the operating status data, fuel consumption data, and location data to the data processing module. The data processing module integrates and analyzes the received operating status data, fuel consumption data, and location data to generate equipment operation indicators. The system acquires the equipment operation indicators in real time through the early warning module, compares the acquired equipment operation indicators with the early warning rules set by the user, and automatically generates and issues early warning information when the early warning conditions are met. The real-time location of the construction equipment and the changing trends of its operational indicators are displayed through a visual interface. The scheduling module receives the early warning information issued by the early warning module, and the management personnel, based on the changing trends of the operational indicators, conduct voice communication with the operators of the construction equipment in the tunnel to execute scheduling instructions.