A shield TBM shield clamping real-time early warning system and early warning method

Abstract

The application discloses a shield TBM (Tunnel Boring Machine) shield clamping real-time early warning system, which comprises a shield TBM shield clamping judgment processing module, a fault-tolerant and safety module, shield TBM shield clamping real-time early warning information, a shield TBM big data platform, a data collector and a field receiving end, wherein the shield TBM shield clamping judgment processing module is electrically connected with the fault-tolerant and safety module, the fault-tolerant and safety module is electrically connected with the shield TBM shield clamping real-time early warning information, the shield TBM shield clamping real-time early warning information is electrically connected with the shield TBM big data platform, the data collector is electrically connected with the shield TBM big data platform, and the shield TBM big data platform is electrically connected with the field receiving end. An early warning method of the early warning system is also disclosed. The application can improve the safety and automation of the algorithm and provide more accurate and timely shield TBM shield clamping real-time early warning for shield construction personnel.

Description

Technical Field

[0001] This invention relates to the field of tunnel boring machine (TBM) construction technology, specifically to a real-time early warning system and method for TBM shield jamming. Background Technology

[0002] A tunnel boring machine (TBM) is the main construction machinery used in tunnel boring machine (TBM) construction. TBM construction is a method of excavating underground tunnels. It uses a TBM to travel underground, preventing the excavation face from collapsing in soft soil or maintaining its stability, and safely carrying out tunnel excavation and lining operations within the machine. TBM tunnel construction is characterized by high automation, labor saving, fast construction speed, one-time excavation, immunity to weather conditions, controllable ground settlement during excavation, minimal impact on surface buildings, and no disruption to surface traffic during underwater excavation. TBM construction is more economical and efficient when dealing with long tunnels and deep burials.

[0003] Due to geological diversity, tunnel boring machine (TBM) construction often faces various risks. Ground subsidence or rock fracturing and compression can cause the TBM to become stuck. Because of the prevalence of TBM sticking, it only attracts attention when it significantly impacts the construction process. Sticking leads to increased energy consumption, severe wear on critical components, slow advance speed, and difficulty in controlling ground subsidence. Timely prediction and appropriate measures can reduce energy consumption, decrease shield wear and material usage, and prevent secondary accidents.

[0004] Traditional methods of judging card blocking rely on manual observation, which are not very accurate and cannot guarantee real-time performance. In particular, when multiple lines are simultaneously being monitored for card blocking, a card blocking early warning system that can quickly respond to multiple lines is needed. Summary of the Invention

[0005] In view of this, the purpose of this invention is to address the shortcomings of existing technologies by providing a real-time early warning system and method for shield tunneling (TBM) shield jamming. This system can achieve real-time sensing and alarm for shield shield shield cutter detachment based on historical data characteristics and real-time data stream changes, combined with the data change characteristics of abnormal cutter detachment. It integrates historical data characteristics, real-time data stream characteristics, and cutter detachment data characteristics to achieve real-time sensing and judgment of independent cutter detachment data from multiple lines. Based on the health status and changes of real-time data, it classifies data changes and performs forward feedback to modify algorithm hyperparameters, enabling the algorithm to have adaptive and self-learning capabilities. Ultimately, this provides shield tunneling personnel with more accurate and timely real-time sensing and alarm for shield shield shield cutter detachment.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A real-time early warning system for shield jamming in a tunnel boring machine (TBM) includes a TBM shield jamming judgment and processing module, a fault tolerance and safety module, real-time early warning information for TBM shield jamming, a TBM big data platform, a data collector, and a field receiver. The TBM shield jamming judgment and processing module is electrically connected to the fault tolerance and safety module, the fault tolerance and safety module is electrically connected to the real-time early warning information for TBM shield jamming, the real-time early warning information for TBM shield jamming is electrically connected to the TBM big data platform, the data collector is electrically connected to the TBM big data platform, and the TBM big data platform is electrically connected to the field receiver.

[0007] The fault tolerance and security module includes a fault tolerance mechanism, an automatic line update check module, an early warning merging module, a criterion update module, and an algorithm security automatic module, which are connected in sequence. The fault tolerance mechanism involves: performing multi-condition judgments on key items of the extracted data; when the ring number key item is invalid, converting the ring number information into mileage information and extracting data by mileage, or converting it into timestamp information and extracting and judging data by time; when the mileage information is invalid, converting the mileage information into ring number information and extracting data by ring number, or converting it into timestamp information and extracting and judging data by time; when feature data items are missing, evaluating the weights of the remaining data items, performing missing item judgment and weighting of the remaining items to ensure the stability and fault tolerance of the algorithm. The automatic line update check: It automatically checks once a day whether the projects under construction have been completed or whether new projects have been added, and automatically completes the replacement after project update and subsequent historical data extraction operations; The warning merging module merges warning information with the same warning interval within 2 hours on the same line with the previous warning information and stores it in the database, only changing the end time, end mileage, and end number change information of the warning. The criterion update module does not completely restrict the criterion parameters for shield jamming in the TBM, ensuring that the criterion parameters and standard values ​​are updated at the criterion layer; at the same time, it sets penalty parameters for the shield jamming perception and early warning module and accepts penalty factor updates generated by real-time data. The algorithm's automatic security module automatically attempts to extract historical data for each loop that does not meet the data requirements, and automatically processes the historical data; it performs daily queries and updates for the lines; it promptly alerts to data anomalies and performs equivalent transformation and fault-tolerant processing to ensure the algorithm's security and automation level.

[0008] Preferably, the shield TBM shield jamming judgment and processing module includes a real-time data processing module and a status and history processing module; The real-time data processing module detects all tunneling lines on the shield tunneling TBM big data platform, extracts the most recent 10-minute historical data, performs equivalent data detection, cleans the real-time data, performs sliding window processing, extracts shield-related feature data, analyzes the feature data, and completes the real-time data processing of one cycle. The status and history processing module checks the tunnel boring machine (TBM) on the big data platform every minute, completes the TBM status query, and queries the TBM type and diameter information; every two rings, it analyzes and cleans the historical data of the project under construction, analyzes the characteristic data related to the shield jamming, and completes the processing of the shield and historical data.

[0009] Preferably, the real-time data processing module includes real-time data accumulation, equivalent data detection, real-time data cleaning, sliding window processing, feature data analysis, and shield tunneling TBM shield jamming early warning criteria, which are connected in sequence. The real-time data accumulation involves extracting the historical data of the previous twelve minutes for each tunnel boring machine in progress, according to the timeframe, to perform initial data accumulation. The equivalent data detection involves: detecting stable and starting segments of the initial accumulated data, detecting time interval clusters, and removing starting segments and time intervals of less than 1 minute. Then, it is determined whether the initial accumulated data meets the 10-minute equivalent data requirement. If not, another beat data is extracted after the initial accumulated data for detection. If the 10-minute equivalent data requirement is met, redundant data at the end is removed. The real-time data cleaning process involves cleaning outliers in the equivalent dataset of the current data to remove any potentially large errors. The sliding window processing involves smoothing the real-time data using a sliding window to obtain information on data trends, fluctuations, and deviation bands. The feature data analysis involves referring to the feature processing methods of historical data and performing similar feature data processing that combines the characteristics of real-time data to obtain the mean, fluctuation, and slope information of the real-time data. The TBM (Tunnel Boring Machine) jamming early warning criteria are as follows: For earth pressure balance (EPB) and slurry balance (SBP) TBMs, in soft soil strata, a 50% decrease in advance speed, a thrust increase exceeding 30%, a torque decrease exceeding 30%, and a 20% increase in articulated cylinder tension are considered critical values ​​indicating shield jamming. For EPB and SBP TBMs, in hard rock strata, an advance speed less than 10 mm / min, a thrust increase exceeding 5000 kN, a torque less than 2000 kN·m, and a 15% increase in articulated cylinder tension are considered critical values ​​indicating shield jamming. For a TBM in soft rock formations, if the advance speed slowly decreases to 10–30 mm / min, the thrust fluctuation range expands by 20% in the early stage and gradually increases in the later stage, the torque decreases by 30%, and the penetration gradually decreases to 3–5 mm / r, then the TBM is considered to have jammed. For a TBM in hard rock formations, if the advance speed decreases by 15%, the thrust fluctuation range increases by 1000 kN, the torque fluctuation range increases by ±1000 kN·m, and the penetration gradually decreases to 3–5 mm / r, then the TBM is considered to have jammed.

[0010] Preferably, the status and history processing module includes shield status query, diameter type query, historical data statistical analysis, historical data cleaning and feature data analysis, which are connected in sequence. The feature data analysis transmits the analysis results to the shield TBM shield jamming early warning criterion. The tunnel boring machine status query: query the tunnel boring machine status information of all lines under construction on the big data platform every minute; The diameter type query: queries the tunneling shield type and diameter information on the big data platform every minute; The historical data statistical analysis involves querying and extracting historical shield tunneling data from all under-construction lines on the big data platform, creating a new thread for each line, and performing tasks such as historical data extraction, equivalent data detection, and data statistical analysis. It also includes distribution analysis of the historical data, including information on data distribution, median, maximum, minimum, and median values. The historical data cleaning process involves: based on historical data analysis, performing absolute cleaning based on engineering data characteristics and relative cleaning based on data continuity; removing outliers based on data distribution characteristics; and checking the continuity of data segments, removing data segments shorter than 1 minute. The feature data analysis involves obtaining information on the median, percentile, mode, and deviation band from the historical data of the shield TBM, including propulsion speed, propulsion force, cutterhead torque, and penetration depth. Smoothing is then performed to obtain information on the slope and data change cycle. The feature data of the historical data, along with its time, ring number, and process history, are stored. Data distribution and health data benchmarks are also acquired and stored to provide a basis for subsequent real-time data analysis.

[0011] Preferably, the TBM shield jamming judgment and processing module reads all historical data information of all lines under construction and all real-time information of all lines in tunneling, and combines them with the TBM shield jamming early warning criteria to make real-time judgments and early warnings. Through the fault tolerance and safety module, a shield jamming real-time early warning system is formed, generating TBM shield jamming real-time early warning information. The shield jamming real-time early warning system transmits the TBM shield jamming real-time early warning information to the TBM big data platform, and then pushes it to the on-site receiving end.

[0012] Preferably, the real-time shield jamming early warning system performs real-time data processing module operations on all under-construction tunneling shield lines to obtain real-time data feature datasets, performs status and history processing module operations on all under-construction lines to obtain information related to the current status, type, diameter, and geology of the shield, and obtains historical data feature datasets. The obtained real-time data feature datasets and historical data feature datasets are compared to perform shield jamming early warning criterion operations on the TBM, the data processing results are delivered to the fault tolerance and safety module, and then real-time alarm for shield jamming of the TBM is performed.

[0013] Preferably, the real-time warning information for TBM jamming is provided separately for earth pressure balance shield tunnels, slurry balance shield tunnels, and TBMs according to their line numbers, and the warning information format is as follows: {Line Number:{Timestamp,Loop Number,Mileage,Card Shield Abnormality:{'Card Shield Number':'Card Shield'},Card Shield Details:{Card Shield Number}}}.

[0014] A method for real-time early warning of shield jamming in a tunnel boring machine (TBM) employs the aforementioned real-time early warning system for shield jamming in TBMs. A data acquisition device collects real-time on-site shield construction data, and a TBM big data platform receives and stores the on-site construction data. On the TBM big data platform, a real-time data processing module extracts and analyzes the real-time data for the corresponding line, while a status and history processing module extracts historical data for the corresponding line, cleans it, and performs feature data analysis to obtain feature data. These features, combined with the current status information of the TBM, are then transmitted to the real-time early warning system for shield jamming in TBMs. The fault tolerance and safety module checks whether key data items are missing and performs corresponding fault tolerance processing, checks line updates, determines whether to merge the same early warning, and finally transmits the processed data to the shield tunneling TBM shield jamming real-time early warning system. The TBM shield jamming real-time early warning system transmits the early warning information to the TBM shield big data platform, which then pushes the information to the on-site receiving end for on-site access.

[0015] The beneficial effects of this invention are: This invention enables real-time early warning of shield jamming for different types of shield TBMs by combining real-time changes in key tunneling parameters such as propulsion speed, propulsion force fluctuations, torque changes, and penetration depth of earth pressure balance shields, slurry balance shields, and tunnel boring machines (TBMs), along with historical construction data and real-time tunneling data. Through multi-threaded concurrency, it establishes a separate data processing and decision-making system for each TBM construction line, integrating multiple individual data anomalies related to shield jamming into a single jamming anomaly event, achieving real-time and rapid judgment of jamming anomalies for different TBM construction lines. It provides corresponding fault tolerance based on the characteristics of current real-time construction data, automatically updates the line status, automatically merges and stores the same early warning information, and allows for the modification of judgment criteria, enabling both algorithmic and manual adjustment. Employing an equivalent execution method and automatic data updates improves algorithm security and automation, providing shield construction personnel with more accurate and timely real-time early warnings of TBM shield jamming. Attached Figure Description

[0016] Figure 1 This is a structural block diagram of the early warning system of the present invention; Figure 2 This is a flowchart of the early warning method of the present invention. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments. Example 1

[0018] like Figure 1 As shown, a real-time early warning system for shield jamming in a tunnel boring machine (TBM) includes a TBM jamming judgment and processing module 1, a fault tolerance and safety module 2, a real-time early warning information module for TBM jamming 3, a TBM big data platform 24, a data collector 23, and a field receiver 25. The TBM jamming judgment and processing module 1 is electrically connected to the fault tolerance and safety module 2, the fault tolerance and safety module 2 is electrically connected to the real-time early warning information module for TBM jamming 3, the real-time early warning information module for TBM jamming 3 is electrically connected to the TBM big data platform 24, the data collector 23 is electrically connected to the TBM big data platform 24, and the TBM big data platform 24 is electrically connected to the field receiver 25.

[0019] The shield-mounted tunnel boring machine (TBM) shield jamming judgment and processing module 1 includes a real-time data processing module 20 and a status and history processing module 21.

[0020] The real-time data processing module 20 includes, in sequence, real-time data accumulation 4, equivalent data detection 5, real-time data cleaning 6, sliding window processing 7, feature data analysis 8, and shield tunneling TBM shield jamming early warning criterion 9, which are electrically connected.

[0021] Real-time data accumulation 4: Extract the historical data of the previous twelve minutes for each tunnel boring machine in the tunneling process according to the rhythm, and perform initial data accumulation.

[0022] Equivalent data detection 5: Perform stable segment and start-up segment detection on the initial accumulated data, and perform time interval point group detection. Remove start-up segment and time interval data with a single segment of less than 1 minute. At this time, determine whether the initial accumulated data meets the 10-minute equivalent data requirement. If it does not meet the requirement, extract another beat data after the initial accumulated data for detection. If it meets the 10-minute equivalent data requirement, remove the redundant data at the end.

[0023] Real-time data cleaning 6: Perform absolute outlier cleaning on the equivalent dataset of the current data to remove potentially gross error values.

[0024] Sliding window processing 7: Perform sliding window smoothing processing on real-time data to obtain data information such as data trend, fluctuation, and deviation band.

[0025] Feature Data Analysis 8: Referring to the feature processing methods of historical data, perform similar feature data processing that combines the characteristics of real-time data to obtain the mean, fluctuation, and slope information of real-time data.

[0026] TBM (Tunnel Boring Machine) Shield Jamming Early Warning Criterion 9: For earth pressure balance (EPB) and slurry balance (SBP) TBMs, in soft soil strata, a 50% decrease in advance speed, a thrust increase exceeding 30%, a torque decrease exceeding 30%, and a 20% increase in articulated cylinder tension are considered critical values ​​indicating shield jamming. For EPB and SBP TBMs, in hard rock strata, an advance speed less than 10 mm / min, a thrust increase exceeding 5000 KN, a torque less than 2000 KN.m, and a 15% increase in articulated cylinder tension are considered critical values ​​indicating shield jamming. For a TBM in soft rock formations, if the advance speed slowly decreases to 10–30 mm / min, the thrust fluctuation range expands by 20% in the early stage and gradually increases in the later stage, the torque decreases by 30%, and the penetration gradually decreases to 3–5 mm / r, then the TBM is considered to have jammed. For a TBM in hard rock formations, if the advance speed decreases by 15%, the thrust fluctuation range increases by 1000 kN, the torque fluctuation range increases by ±1000 kN·m, and the penetration gradually decreases to 3–5 mm / r, then the TBM is considered to have jammed.

[0027] The real-time data processing module 20 detects all tunneling lines on the shield tunneling TBM big data platform 24, extracts the most recent 10-minute historical data, performs equivalent data detection, cleans the real-time data, performs sliding window processing, extracts shield-related feature data, analyzes the feature data, and completes the real-time data processing for one cycle.

[0028] The status and history processing module 21 includes shield status query 10, diameter type query 11, historical data statistical analysis 12, historical data cleaning 13 and feature data analysis 8, which are connected in sequence. The feature data analysis 8 transmits the analysis results to the shield TBM shield jamming early warning criterion 9.

[0029] Tunnel Boring Machine Status Query 10: Query tunnel boring machine status information for all tunnels under construction on the big data platform every minute.

[0030] Diameter Type Query 11: Query the type and diameter information of tunnel boring machines on the big data platform every minute.

[0031] Historical Data Statistical Analysis 12: Query and extract historical shield tunneling data for all tunnels under construction on the big data platform, open a new thread for each tunnel, and perform tasks such as historical data extraction, equivalent data detection, and data statistical analysis; and perform distribution analysis on the historical data, including data distribution analysis, median, maximum value, minimum value, and related information.

[0032] Historical data cleaning 13: Based on the analysis of historical data, perform absolute cleaning based on the characteristics of engineering data and relative cleaning based on the continuity of data. Remove isolated points based on the data distribution characteristics and check the continuity of data segments. Remove data segments shorter than 1 minute.

[0033] Feature Data Analysis 8: Obtain the median, percentile, mode, and deviation band information from the historical data of the shield TBM, including propulsion speed, propulsion force, cutterhead torque, and penetration depth. Perform smoothing processing to obtain information on slope and data change cycle. Store the feature data of historical data and related information such as time, ring number, and process history. Obtain and store the data distribution and health data benchmark to provide a basis for subsequent real-time data judgment.

[0034] The status and history processing module 21 checks the tunnel boring machine (TBM) on the big data platform every minute, completes the TBM status query, and queries the TBM type and diameter information; every two rings, it analyzes and cleans the historical data of the project under construction, analyzes the characteristic data related to the shield jamming, and completes the processing of the shield and historical data.

[0035] The fault tolerance and security module 2 includes a fault tolerance mechanism 15, an automatic line update check 16, an early warning merging module 17, a criterion updateable module 18, and an algorithm security automatic module 19, which are connected in sequence.

[0036] Fault Tolerance Mechanism 15: Multiple conditions are applied to key data extraction items. When the ring number key item fails, the ring number information is converted to mileage information for extraction, or converted to timestamp information for extraction and judgment based on time. When mileage information fails, the mileage information is converted to ring number information for extraction, or converted to timestamp information for extraction and judgment based on time. When feature data items are missing, the weights of the remaining data items are evaluated, and missing item judgment and remaining item weighting are performed to ensure algorithm stability and fault tolerance.

[0037] Automatic Line Update Check 16: Automatically check once a day whether projects under construction have been completed or whether new projects have been added, and automatically complete the replacement of updated projects and subsequent historical data extraction.

[0038] Warning merging module 17: For warning information with the same warning interval within 2 hours on the same line, merge it with the previous warning information and store it in the database, only changing the relevant information such as the end time, end mileage, and end number change of the warning.

[0039] Criterion Update Module 18: It does not impose complete restrictions on the criterion parameters of the shield TBM stuck shield, ensuring that the criterion parameters and standard values ​​can be updated at the criterion layer; at the same time, it sets penalty parameters for the shield stuck shield perception and early warning module, and accepts the penalty factor update generated by real-time data.

[0040] Algorithm Security Automated Module 19: For lines that do not meet data requirements, it automatically attempts to extract historical data once for each loop and automatically completes the processing of historical data; it performs daily query updates for lines; it promptly alerts for data anomalies and performs equivalent transformation and fault tolerance processing for data anomalies to ensure the security and automation level of the algorithm.

[0041] The real-time early warning information for TBM (Tunnel Boring Machine) jamming is as follows: Early warnings are issued for earth pressure balance shield tunnels, slurry balance shield tunnels, and TBMs according to their line numbers. The format of the early warning information is as follows: {Line Number:{Timestamp,Loop Number,Mileage,Card Shield Abnormality:{'Card Shield Number':'Card Shield'},Card Shield Details:{Card Shield Number}}}.

[0042] The following are examples of card shield anomalies: {'21090003000301':{'time':1659888225,'ringNum':411,'mileage':52352.45,'exception' :{'4a':'Shield'},'exceptionParameter':{'4a':'p1:29.523,p2:0.221,p3:0.124,p4:3.792'}}}.

[0043] The shield TBM shield jamming judgment and processing module 1 reads all historical data information of all lines under construction and all real-time information of all lines under tunneling, and combines it with the shield TBM shield jamming early warning criterion 9 to make real-time judgments and early warnings. Through the fault tolerance and safety module 2, a shield jamming real-time early warning system 22 is formed, which generates shield TBM shield jamming real-time early warning information 3. The shield TBM shield jamming real-time early warning system 22 transmits the shield TBM shield jamming real-time early warning information 3 to the shield TBM big data platform 24, and then pushes it to the field receiving end 25.

[0044] The shield jamming real-time early warning system 22 operates the real-time data processing module 20 to obtain real-time data feature datasets for all tunneling shield lines under construction, and operates the status and history processing module 21 to obtain information related to the current status, type, diameter, and geology of the shield for all lines under construction, and obtains the feature dataset of historical data. The obtained real-time data feature dataset and the historical data feature dataset are compared to perform the shield jamming early warning criterion 9 operation, and the data processing results are delivered to the fault tolerance and safety module 2, and then the shield jamming real-time alarm is performed. Example 2

[0045] A real-time early warning method for shield jamming in a tunnel boring machine (TBM) employs a data acquisition device 23 to collect real-time on-site shield construction data. A TBM big data platform 24 receives and stores the on-site construction data. The TBM big data platform 24 utilizes two Intel(R) Xeon(R) CPU E5-2650 v4@2.20GHz processors with 128GB of memory as a cache server; two Intel(R) Xeon(R) CPU E5-2698R v4@2.20GHz server processors with 128GB of memory and an 80TB hard disk storage cluster as a data storage server; and two Intel(R) Xeon(R) CPU E5-2650 v4@2.20GHz processors with 128GB of memory as a network publishing server.

[0046] The real-time data processing module 20 extracts and analyzes the real-time data of the corresponding line, while the status and history processing module 21 extracts the historical data of the corresponding line, cleans it, and performs feature data analysis to obtain feature data. The features are combined with the current status information of the shield TBM and transmitted to the shield TBM real-time early warning system 22.

[0047] The fault tolerance and safety module 2 checks whether key data items are missing and performs corresponding fault tolerance processing, checks line updates, determines whether to merge the same early warning, and finally transmits the processed data to the shield tunneling TBM real-time early warning system 22.

[0048] The TBM (Tunnel Boring Machine) real-time early warning system 22 transmits the early warning information to the TBM big data platform 24, which then pushes the information to the field receiver 25 for on-site access. The field receiver 25 uses two Intel(R) Xeon(R) CPU E5-2650 v4@2.20GHz processors and is equipped with 128GB of memory.

[0049] This invention can provide real-time early warning of shield jamming for different types of shield TBMs by combining real-time changes in key tunneling parameters such as the advance speed, thrust fluctuation, torque variation, and penetration of earth pressure balance shields, slurry balance shields, and TBMs with historical construction data and real-time tunneling data.

[0050] By assessing the construction status of the tunnel boring machine (TBM) and the tunneling progress of the tunnel, a separate data processing and decision-making system is established for each TBM construction line through multi-threaded concurrency. This system integrates multiple individual data anomalies related to shield jamming into a single shield jamming event, enabling real-time and rapid assessment of shield jamming anomalies that are adaptable to different TBM construction lines.

[0051] Based on the characteristics of the current real-time construction data, corresponding fault tolerance measures are implemented. The line status is checked and the line is automatically updated. The same warning information is automatically merged, stored, and displayed. The criteria are set to be modifiable, so that the algorithm and human adjustment of the criteria can be achieved. An equivalent execution method is adopted, and automatic data updates are implemented to improve the security and automation of the algorithm, providing shield tunneling workers with more accurate and timely real-time warnings of shield jamming.

[0052] Those skilled in the art will understand that the embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0053] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0054] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0055] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0056] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are only illustrative of the principles of the present invention. Various changes and modifications can be made to the present invention without departing from the spirit and scope of the present invention, and all such changes and modifications fall within the scope of the present invention as claimed.

Claims

1. A real-time early warning system for shield jamming in a tunnel boring machine (TBM), characterized in that: The system includes a TBM shield jamming judgment and processing module (1), a fault tolerance and safety module (2), a TBM shield jamming real-time early warning information (3), a TBM big data platform (24), a data collector (23), and a field receiver (25). The TBM shield jamming judgment and processing module (1) is electrically connected to the fault tolerance and safety module (2), the fault tolerance and safety module (2) is electrically connected to the TBM shield jamming real-time early warning information (3), the TBM shield jamming real-time early warning information (3) is electrically connected to the TBM big data platform (24), the data collector (23) is electrically connected to the TBM big data platform (24), and the TBM big data platform (24) is electrically connected to the field receiver (25). The fault tolerance and security module (2) includes a fault tolerance mechanism (15), an automatic line update check (16), an early warning merging module (17), a criterion updateable module (18), and an algorithm security automatic module (19) connected in sequence. The fault tolerance mechanism (15) performs multi-condition judgment on the key items of the extracted data. When the ring number key item fails, the ring number information is converted into mileage information and extracted according to mileage, or converted into timestamp information and extracted and judged according to time. When the mileage information fails, the mileage information is converted into ring number information and extracted according to ring number, or converted into timestamp information and extracted and judged according to time. When the card shield related feature data is missing, the weight of the remaining data items is evaluated, the missing item judgment and the remaining item weighting are performed to ensure the stability and fault tolerance of the algorithm. The card shield related feature data includes propulsion speed, thrust, torque, articulated cylinder pull force, and penetration. The automatic line update check (16) automatically checks once a day whether the projects under construction have been completed and whether there are any newly added projects, and automatically completes the replacement and subsequent historical data extraction operations after the project update. The warning merging module (17) merges warning information with the previous warning information for the same line within 2 hours into the database, and only changes the end time, end mileage, and end number change information of the warning. The criterion update module (18) does not impose complete restrictions on the criterion parameter items of the shield TBM shield jamming, ensuring that the criterion parameter items and standard values ​​are updated at the criterion layer; at the same time, it sets penalty item parameters for the shield jamming perception and early warning module and accepts penalty factor updates generated by real-time data. The algorithm security automatic module (19) automatically attempts to extract historical data once for each loop that does not meet the data requirements, and automatically completes the processing of historical data; it performs daily query updates for the lines; it promptly alarms for data anomalies, and performs equivalent transformation and fault tolerance processing for data anomalies to ensure the security and automation of the algorithm.

2. The real-time early warning system for shield jamming in a tunnel boring machine (TBM) according to claim 1, characterized in that: The shield tunneling machine (TBM) shield jamming judgment and processing module (1) includes a real-time data processing module (20) and a status and history processing module (21); The real-time data processing module (20) detects all tunneling lines on the shield tunneling TBM big data platform (24), extracts 10 minutes of historical data from the most recent time period, performs equivalent data detection, cleans the real-time data, performs sliding window processing, extracts shield-related feature data and analyzes the feature data, and completes one cycle of real-time data processing. The status and history processing module (21) checks the tunnel boring machine (TBM) that is tunneling on the big data platform once per minute, completes the TBM status query, and queries the TBM type and diameter information; every two rings, it analyzes the historical data of the project under construction and cleans the data, analyzes the characteristic data related to the shield, and completes the processing of the shield and historical data.

3. The real-time early warning system for shield jamming in a tunnel boring machine (TBM) according to claim 2, characterized in that: The real-time data processing module (20) includes real-time data accumulation (4), equivalent data detection (5), real-time data cleaning (6), sliding window processing (7), feature data analysis (8), and shield TBM shield jamming early warning criterion (9) connected in sequence. The real-time data accumulation (4): The history of the previous twelve minutes for each tunnel boring machine in the tunneling process is extracted according to the beat, and the data is initially accumulated; The equivalent data detection (5): The initial accumulated data is tested for stable segments and starting segments, and the time interval point group is tested. The starting segment and the time interval data of a single segment less than 1 minute are removed. At this time, it is determined whether the initial accumulated data meets the 10-minute equivalent data requirement. If it does not meet the requirement, another beat data is extracted after the initial accumulated data for detection. If it meets the 10-minute equivalent data requirement, the redundant data at the end is removed. The real-time data cleaning (6) involves cleaning outliers in the equivalent data set of the current data to remove any gross error values ​​that may occur. The sliding window processing (7) involves smoothing the real-time data using a sliding window to obtain information on data trends, fluctuations, and deviation bands. The feature data analysis (8): Referring to the feature processing method of historical data, similar feature data processing is performed in combination with the characteristics of real-time data to obtain the mean, fluctuation and slope information of real-time data; The shield TBM jamming early warning criterion (9): For earth pressure balance shields and slurry balance shields, in soft soil strata, if the advance speed decreases by 50%, the thrust increases by more than 30%, the torque decreases by more than 30%, and the articulated cylinder pull increases by 20%, then the shield jamming is judged as a critical value; For earth pressure balance shields and slurry balance shields, in hard rock strata, if the advance speed is less than 10 mm / min, the thrust increases by more than 5000 KN, and the torque is less than 2000 KN, then the shield jamming is judged as a critical value; If the tension of the articulated hydraulic cylinder increases by 15%, it is considered a critical value indicating shield jamming. For TBMs in soft rock formations, if the advance speed slowly decreases to 10–30 mm / min, the thrust fluctuation range expands by 20% in the early stage and gradually increases in the later stage, the torque decreases by 30%, and the penetration gradually decreases to 3–5 mm / r, it is considered a critical value indicating shield jamming. For TBMs in hard rock formations, if the advance speed decreases by 15%, the thrust fluctuation range increases by 1000 kN, the torque fluctuation range increases by ±1000 kN·m, and the penetration gradually decreases to 3–5 mm / r, it is considered a critical value indicating shield jamming.

4. The real-time early warning system for shield jamming in a tunnel boring machine (TBM) according to claim 3, characterized in that: The status and history processing module (21) includes shield status query (10), diameter type query (11), historical data statistical analysis (12), historical data cleaning (13) and feature data analysis (8) connected in sequence. The feature data analysis (8) transmits the analysis results to the shield TBM shield jamming early warning criterion (9). The shield tunneling status query (10) queries the shield tunneling status information of all lines under construction on the big data platform every minute; The diameter type query (11): queries the type and diameter information of tunneling shields on the big data platform every minute; The historical data statistical analysis (12): query and extract the shield tunneling historical data information of all lines under construction on the big data platform, open up a new thread for each line, and carry out the tasks of historical data extraction, equivalent data detection and data statistical analysis. And perform distribution analysis on historical data, including data distribution analysis, median, maximum value, minimum value, and median information; the historical data cleaning (13): based on the historical data analysis, perform absolute cleaning based on engineering data characteristics and relative cleaning based on data continuity, remove isolated points based on data distribution characteristics, check the continuity of data segments, and remove data segments less than 1 minute. The feature data analysis (8) involves obtaining the median, percentile, mode, and deviation band information from the historical data of the shield TBM, including propulsion speed, propulsion force, cutterhead torque, and penetration depth. It also involves smoothing the data to obtain information on the slope and data change cycle, storing the feature data of the historical data, and related information on its time, ring number, and process. The analysis also involves obtaining and storing the data distribution and health data benchmark to provide a basis for subsequent real-time data judgment.

5. The real-time early warning system for shield jamming in a tunnel boring machine (TBM) according to claim 1, characterized in that: The shield TBM shield jamming judgment and processing module (1) reads all historical data information of all lines under construction and all real-time information of all tunneling lines, and combines them with the shield TBM shield jamming early warning criteria (9) to make real-time judgment and early warning. Through the fault tolerance and safety module (2), a shield jamming real-time early warning system (22) is formed, generating shield TBM shield jamming real-time early warning information (3). The shield jamming real-time early warning system (22) transmits the shield TBM shield jamming real-time early warning information (3) to the shield TBM big data platform (24), and then pushes it to the on-site receiving end (25).

6. The real-time early warning system for shield jamming in a tunnel boring machine (TBM) according to claim 5, characterized in that: The real-time shield jamming early warning system (22) operates the real-time data processing module (20) to obtain the real-time data feature dataset for all tunneling shield lines under construction, operates the status and history processing module (21) to obtain the current status, type, diameter and geological information of the shield, and obtain the feature dataset of historical data. The system compares the obtained real-time data feature dataset and the historical data feature dataset to perform the shield TBM shield jamming early warning criterion (9) operation, delivers the data processing results to the fault tolerance and safety module (2), and then performs real-time shield TBM shield jamming alarm.

7. The real-time early warning system for shield jamming in a tunnel boring machine (TBM) according to claim 1, characterized in that: The real-time early warning information (3) for shield TBM jamming provides early warnings for earth pressure balance shield, slurry balance shield and TBM according to the line number. The format of the early warning information is as follows: {line number:{timestamp, ring number, mileage, shield jamming abnormality:{'shield jamming number':'shield jamming'}, shield jamming details:{shield jamming number}}}.

8. A method for real-time early warning of shield jamming in a tunnel boring machine (TBM), employing the real-time early warning system for shield jamming in a TBM as described in any one of claims 1-7, characterized in that: The data acquisition unit (23) collects on-site shield tunneling construction data in real time. The shield TBM big data platform (24) receives on-site construction data and stores it. On the shield TBM big data platform (24), the real-time data processing module (20) extracts the real-time data of the corresponding line and analyzes and processes it. The status and history processing module (21) extracts the historical data of the corresponding line and performs cleaning and feature data analysis to obtain feature data. The feature data is combined with the current status information of the shield TBM and transmitted to the shield TBM shield jamming real-time early warning system (22). The fault tolerance and safety module (2) checks whether key data items are missing and performs corresponding fault tolerance processing, checks line updates, determines whether to merge the same early warning, and finally transmits the processed data to the shield TBM card shield real-time early warning system (22). The shield tunneling machine (TBM) real-time early warning system (22) transmits the early warning information to the shield tunneling machine big data platform (24), which then pushes the information to the on-site receiver (25) for on-site access.

Images