Intelligent scheduling system for tunnel construction operation objects

Abstract

The invention discloses an intelligent scheduling system for tunnel construction operation objects, and the system is characterized in that the construction system comprises a data collection module, a data processing module, a scheduling decision module, an execution and feedback module, and a user interface module; through highly integrated data acquisition, processing and decision making modules, the system can respond to the change of a construction site in real time and optimize scheduling decisions, so that the construction efficiency and the resource utilization rate are remarkably improved. Through the real-time monitoring and dynamic adjustment functions of the execution and feedback module, the scheduling system can respond to onsite emergencies in time, project delay and overhigh cost are reduced, and the construction progress and quality are guaranteed. The user interface module provides a visual and easy-to-operate interaction platform, so that a dispatcher can effectively monitor and manage the whole construction process, and the operation convenience and the decision accuracy are improved. The risk assessment and management function of the system strengthens the construction safety, and by predicting and preventing potential risks, the accident rate is obviously reduced, and the safety of personnel and equipment is guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of building construction management, and in particular to an intelligent scheduling system for tunnel construction operation objects. Background Art

[0002] In modern tunnel construction projects, an efficient scheduling system is one of the key factors to ensure that the project is completed on time and operates safely. Traditional tunnel construction scheduling mostly relies on manual operations, which is not only inefficient, but also prone to errors in complex and changing construction environments, increasing the risk of accidents. Tunnel construction involves a large number of mechanical equipment and personnel working together in a narrow and ever-changing space. Any scheduling errors may lead to serious safety accidents and increased costs. In addition, the tunnel construction environment is complex and the working conditions are extremely challenging, including harsh climatic conditions, terrain restrictions, and uncertain underground environments. These factors have greatly increased the difficulty of construction scheduling. Existing scheduling methods often fail to respond quickly and effectively due to the lack of real-time monitoring and data analysis support, and cannot update construction strategies in real time, thus affecting overall construction efficiency and safety management.

[0003] The existing tunnel construction scheduling technology has the following main disadvantages:

[0004] Slow response and inefficiency: Due to the lack of real-time data support, the existing technology has a long scheduling response time and cannot quickly adapt to on-site changes, resulting in low construction efficiency.

[0005] Insufficient safety management: Manual scheduling is highly dependent and error-prone, and lacks effective risk prediction and accident prevention mechanisms, increasing the risk of safety accidents.

[0006] Unreasonable resource allocation: Due to the lack of accurate data analysis, resource allocation is often not scientific enough, resulting in waste of manpower and mechanical resources.

[0007] Although the existing system plays a basic scheduling role in tunnel construction projects, its efficiency and safety still need to be improved. In particular, it lacks flexibility and foresight when dealing with emergencies and complex scenarios. These shortcomings are particularly evident in high-risk tunnel construction environments, seriously affecting project progress and quality.

[0008] Therefore, developing an intelligent system that can monitor the construction site in real time and automatically make scheduling decisions has become an important requirement for improving the efficiency and safety of tunnel construction.

[0009] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0010] The purpose of the present invention is to provide an intelligent scheduling system for tunnel construction work objects to solve the problems existing in the prior art.

[0011] In order to achieve the above object, the present invention adopts the following technical solutions:

[0012] The present invention provides an intelligent scheduling system for tunnel construction objects, the construction system comprising: a data acquisition module, a data processing module, a scheduling decision module, an execution and feedback module and a user interface module; wherein,

[0013] The data acquisition module is used to capture the precise location and operating status of personnel and vehicles on the tunnel construction site in real time, and to format the collected data into standard longitude and latitude coordinates and specific status descriptions, and transmit them to the data processing module in real time for storage and backup;

[0014] The data processing module is used to conduct in-depth analysis and processing of a large amount of raw data received from the data acquisition module, including data cleaning and preprocessing, feature extraction, behavior analysis and risk assessment, and send the analyzed and processed data to the scheduling decision module to ensure that the scheduling decision module can respond based on accurate and real-time information;

[0015] The scheduling decision module is used to convert the collected and processed data into an actual executable scheduling plan, and send the scheduling plan to the execution and feedback module; including task priority allocation, resource allocation and path planning;

[0016] The execution and feedback module is used to convert the scheduling plan received from the scheduling decision module into specific operation instructions, and send them to the personnel and mechanical equipment on site through the communication system;

[0017] The user interface module is used to directly display the real-time data collected by the data acquisition module. These real-time data are analyzed and organized by the data processing module and converted into information that can be directly used by users. This information is intuitively displayed in the form of charts, indicators and maps on the interface, so that dispatchers can quickly understand the on-site situation; the scheduling plans and instructions generated by the scheduling decision module are also transmitted to the operators through the user interface. The interface provides operation functions, and the dispatchers make decisions and adjustments based on real-time data and system recommendations; the execution results and status feedback collected by the execution and feedback module are also updated in real time on the user interface, helping dispatchers monitor the progress and effects of tasks and make necessary adjustments in a timely manner.

[0018] Further, the process of data cleaning and preprocessing is as follows:

[0019] Remove noise and outliers: First, use the Z-Score method to standardize the data; for each data point, calculate its deviation from the mean and divide it by the standard deviation. The resulting Z value represents the standard score of the data point; the formula is:

[0020] ;

[0021] in is the original data point, is the average value, is the standard deviation; when When the absolute value of is greater than 3, the data point is considered an outlier and is removed;

[0022] Data smoothing: Kalman filtering algorithm is applied to smooth the position data to eliminate random fluctuations caused by equipment errors or environmental factors. Kalman filtering optimizes the accuracy of data through two steps: prediction and update. Its basic formula is:

[0023] ;

[0024] ;

[0025] in is the current estimated state, is the actual measured value, is the observation matrix, is the Kalman gain, is the measurement noise covariance matrix, is the forecast error covariance.

[0026] Further, the feature extraction process is as follows:

[0027] Feature extraction is to extract useful information from preprocessed data to provide input for subsequent behavior analysis and risk assessment; in the tunnel construction scheduling system, key features include speed, acceleration and dwell time;

[0028] Velocity and acceleration calculations:

[0029] speed It is calculated directly from the location information provided by GPS according to the time change; the formula is:

[0030] ;

[0031] in, is the position change, is the time variation;

[0032] Acceleration is the time derivative of velocity, calculated from two consecutive velocity measurements:

[0033] ;

[0034] in, is the velocity change;

[0035] Duration:

[0036] For a specific area, the residence time refers to the total time a vehicle or person spends in the area; by analyzing the location time series data, when the timestamps of entering and leaving a specific area are recorded, the residence time is the difference between these two timestamps.

[0037] Further, the process of the behavior analysis is as follows:

[0038] Behavioral analysis using SVM:

[0039] Leverage extracted data including vehicle speed, acceleration, and dwell time in specific areas; define the types of behavior that need to be identified, including idling, moving, or loading activities; label the data accordingly using manual input from historical observations or expert insights;

[0040] Model training: Use labeled data sets to train SVM; SVM will learn to distinguish different behaviors based on input features; including selecting kernel type and adjusting parameters, regularization parameters and kernel coefficients, the formula is as follows:

[0041] ;

[0042] in, is the input feature, is the category label, are the learned parameters, is the kernel function, is the bias term;

[0043] Cross-validation: Use k-fold cross-validation technique to evaluate the performance of SVM to ensure that the model has good generalization ability to unseen data; adjust parameters based on accuracy, precision, recall and F1 score;

[0044] Behavior prediction: Once training and validation are complete, deploy the SVM model to predict the behavior of new data in real time. The model will classify vehicle and human activities based on current speed, acceleration, and other contextual data.

[0045] Output: The final output of the SVM analysis will be the classification of each observation instance into the predefined behavioral categories.

[0046] Further, the risk assessment process is as follows:

[0047] Risk assessment is to predict potential security risks based on the extracted features and behavior analysis results. It includes the following steps:

[0048] Step 1: Define risk indicators:

[0049] First determine which behaviors or conditions are high risk;

[0050] Step 2: Collect Data:

[0051] According to the defined risk indicators, corresponding data is collected from feature extraction and behavior analysis modules;

[0052] Step 3: Risk Calculation:

[0053] A logistic regression model is applied to assess risk; the model takes the characteristics of various behaviors and states as input and outputs a risk probability value, calculated as:

[0054] ;

[0055] in, is the probability of the risk occurring. are input features, are model parameters;

[0056] Step 4: Risk level classification:

[0057] Classify behaviors into different risk levels based on risk probability; and set thresholds to distinguish these levels;

[0058] Step 5: Report and Respond:

[0059] The risk assessment results are reported to the dispatch center, and corresponding prevention or intervention measures are taken according to the risk level;

[0060] Based on the extracted features and behavior analysis results, a logistic regression model is used to evaluate potential security risks; the model calculates the risk probability using the following formula:

[0061] ;

[0062] in is the probability of the event occurring, are model parameters, It is an influencing factor;

[0063] Through the above comprehensive data processing strategy, the intelligent scheduling system can effectively extract valuable information from huge data sets and provide scientific and accurate support for scheduling decisions.

[0064] Furthermore, the specific process of the path planning is as follows:

[0065] The Dijkstra algorithm is used for path planning, as follows:

[0066] Step 1: Initialization:

[0067] Set the distance of all nodes to infinity and the distance of the starting point to 0;

[0068] Create a priority queue to store each node and its distance;

[0069] Step 2: Algorithm Execution:

[0070] Take the node with the smallest distance from the priority queue and update the distance of its adjacent nodes;

[0071] For the current node And every adjacent node , if through arrive The distance is less than the current record Distance, update distance, and Add to the priority queue; the formula is:

[0072] ;

[0073] in From the starting point to the node The shortest distance, Is a node arrive The distance of the edge;

[0074] Step 3: Iterative Update:

[0075] Repeat the above steps until the priority queue is empty, at which point the shortest paths to all reachable nodes have been found;

[0076] Step 4: Result extraction:

[0077] Finally, the shortest distance from the starting point to each node will be stored in the distance array middle;

[0078] Through the Dijkstra algorithm, the scheduling system can efficiently plan the shortest path from any designated starting point to all key points in the construction area, ensuring the optimal efficiency of resource allocation and personnel mobilization.

[0079] By adopting the above technical solution, the present invention has the following beneficial effects:

[0080] Improve dispatch response speed and efficiency: Through real-time data monitoring and automated dispatch algorithms, the response time can be significantly shortened and the efficiency and accuracy of dispatch decisions can be improved.

[0081] Enhance safety management capabilities: Integrate risk assessment and accident prevention mechanisms, monitor potential risks in real time, adjust construction strategies in a timely manner, and reduce the accident rate.

[0082] Optimize resource allocation: Use data analysis to optimize the allocation of personnel and machinery, reduce resource waste, ensure that resources are effectively used where necessary, and improve overall construction efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0084] Figure 1 A structural diagram of an intelligent scheduling system for tunnel construction work objects provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0085] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0086] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0087] Combination Figure 1 As shown, the embodiment of the present invention provides a comprehensive solution tailored for tunnel construction projects, the purpose of which is to optimize the scheduling efficiency and safety of personnel and vehicles through real-time data collection and processing. The system includes a data collection module, a data processing module, a scheduling decision module, an execution and feedback module, and a user interface module. The details are as follows:

[0088] 1. Data acquisition module:

[0089] The data acquisition module is the foundation of the intelligent dispatching system. It is responsible for capturing the precise location and operating status of personnel and vehicles on the tunnel construction site in real time. The module uses high-precision GPS technology to track the location of vehicles and personnel in outdoor activities, and combines RFID technology to monitor the dynamics indoors or inside the tunnel. In addition, the module also collects information about the vehicle's running speed, direction, and mechanical operating status through various sensors. These data are formatted into standard latitude and longitude coordinates and specific status descriptions, and transmitted to the central database in real time for storage and backup to ensure data integrity and reliability.

[0090] 2. Data processing module:

[0091] The data processing module is a key component in the intelligent scheduling system, responsible for in-depth analysis and processing of the large amount of raw data received from the data acquisition module. The main tasks of this module include data cleaning, feature extraction, behavior analysis and risk assessment to ensure that the scheduling decision module can respond based on accurate and real-time information.

[0092] (1) Data cleaning and preprocessing:

[0093] Remove noise and outliers: First, use the Z-Score method to standardize the data. For each data point, calculate its deviation from the mean and divide it by the standard deviation. The resulting Z value represents the standard score of the data point. The formula is:

[0094] ;

[0095] in is the original data point, is the average value, is the standard deviation. In general, when When the absolute value of is greater than 3, the data point is considered an outlier and is removed.

[0096] Data smoothing: Apply the Kalman filter algorithm to smooth the position data to eliminate random fluctuations caused by equipment errors or environmental factors. Kalman filtering optimizes the accuracy of data through two steps: prediction and update. Its basic formula is:

[0097] ;

[0098] ;

[0099] in is the current estimated state, is the actual measured value, is the observation matrix, is the Kalman gain, is the measurement noise covariance matrix, is the forecast error covariance.

[0100] (2) Feature extraction:

[0101] It is responsible for extracting useful information from the preprocessed data and providing input for subsequent behavior analysis and risk assessment. In the tunnel construction scheduling system, key features usually include speed, acceleration, and dwell time.

[0102] Velocity and acceleration calculations:

[0103] speed It can be directly calculated based on the time change of the location information provided by GPS. The formula is:

[0104] ;

[0105] in, is the position change, is the time variation.

[0106] Acceleration is the time derivative of velocity and can be calculated from two consecutive velocity measurements:

[0107] ;

[0108] in, is the speed change.

[0109] Duration:

[0110] For a specific area, the residence time refers to the total time a vehicle or person spends in the area. By analyzing the location time series data, when the timestamps of entering and leaving a specific area are recorded, the residence time is the difference between these two timestamps.

[0111] Behavioral analysis:

[0112] Behavioral analysis using SVM

[0113] Leverage previously extracted data such as vehicle speed, acceleration, and dwell time in specific areas. These features are able to capture key factors that influence the operation patterns at the construction site. Define the types of behaviors that need to be identified, such as idling, moving, or loading activities. Label the data accordingly, requiring manual input from historical observations or expert insights to ensure accuracy.

[0114] Model training: Use the labeled dataset to train the SVM. The SVM will learn to distinguish different behaviors based on the input features. This includes choosing the kernel type (such as linear, polynomial, or radial basis function) and adjusting parameters such as the regularization parameter (C) and kernel coefficient.

[0115] ;

[0116] in, is the input feature, is the category label, are the learned parameters, is the kernel function, is the bias term.

[0117] Cross-validation: Use techniques such as k-fold cross-validation to evaluate the performance of the SVM to ensure that the model has good generalization ability to unseen data. Adjust parameters based on performance indicators such as accuracy, precision, recall, and F1 score.

[0118] Behavior prediction: Once training and validation are complete, the SVM model is deployed to predict the behavior of new data in real time. The model will classify vehicle and human activities based on current speed, acceleration, and other contextual data.

[0119] Output: The final output of the SVM analysis will be the classification of each observation instance into the predefined behavioral categories.

[0120] risk assessment:

[0121] Risk assessment is based on the extracted features and behavior analysis results to predict potential security risks.

[0122] Step 1: Define risk indicators

[0123] First determine which behaviors or conditions are high risk, such as speeding, unusually long stops, proximity to dangerous areas, etc.

[0124] Step 2: Collect Data

[0125] Based on the defined risk indicators, corresponding data is collected from the feature extraction and behavior analysis modules. For example, the speed and dwell time data of each vehicle in a specific area are collected.

[0126] Step 3: Risk Calculation

[0127] A logistic regression model is applied to assess risk. The model takes the characteristics of various behaviors and states as input and outputs a risk probability value, which is calculated as:

[0128]

[0129] in, is the probability of the risk occurring. are input features, are model parameters.

[0130] Step 4: Risk level classification

[0131] Behaviors are classified into different risk levels, such as low risk, medium risk, and high risk, based on the risk probability. Thresholds can be set to distinguish these levels, for example, a probability below 0.3 is low risk, 0.3 to 0.7 is medium risk, and above 0.7 is high risk.

[0132] Step 5: Report and respond

[0133] The risk assessment results are reported to the dispatch center, and corresponding prevention or intervention measures are taken according to the risk level. For example, for high-risk behaviors, it may be necessary to immediately adjust the dispatch plan or take safety measures.

[0134] Based on the extracted features and behavior analysis results, a logistic regression model is used to evaluate potential security risks. The model calculates the risk probability using the following formula:

[0135]

[0136] in is the probability of the event occurring, are model parameters, are influencing factors (such as speed and distance).

[0137] Through this comprehensive data processing strategy, the intelligent scheduling system can effectively extract valuable information from huge data sets and provide scientific and accurate support for scheduling decisions.

[0138] 3. Scheduling decision module:

[0139] The scheduling decision module is the "brain" of the intelligent scheduling system. It is responsible for converting the collected and processed data into an actual executable scheduling plan. The core functions of this module include task priority allocation, resource allocation and path planning.

[0140] First, the module integrates the previous risk assessment results and behavior analysis data, such as vehicle behavior patterns, risk levels, and regional congestion. The decision module prioritizes tasks based on their urgency, relevance, and resource requirements, giving higher priority to tasks with high safety requirements or strict time constraints.

[0141] Next, resource allocation is carried out. The system will conduct a detailed assessment of the resource requirements for each task, including the required manpower, machine type and quantity, and time required. Next, the system will check the resource inventory to determine the currently available resources. This includes the skill level of the personnel, the availability of the machinery (such as maintenance and repair status), and location information. When allocating resources, the system will consider the urgency and priority of the task. For critical tasks that affect the progress of the entire project, the best available resources will be allocated first.

[0142] This module uses advanced optimization algorithms, such as genetic algorithms or linear programming, to determine how to most effectively allocate resources and plan routes to maximize work efficiency and minimize costs and time delays. Dispatch instructions range from simple vehicle dispatch to complex multi-resource coordination tasks. In order to improve the flexibility and responsiveness of the system, this module also allows dispatchers to manually intervene and adjust according to real-time conditions to ensure that dispatch instructions can adapt to emergencies. The system will use advanced algorithms to resolve conflicts in resource allocation and ensure that resources do not conflict between multiple tasks. At the same time, the algorithm will also try to find the most optimized resource configuration to maximize resource utilization efficiency.

[0143] Finally, path planning is carried out. In the intelligent scheduling system, the Dijkstra algorithm is used to calculate the shortest path from one point on the construction site to other points, which is particularly suitable for path planning in dynamic environments.

[0144] 1. Initialization:

[0145] Set the distance of all nodes to infinity and the distance of the starting point to 0.

[0146] Create a priority queue (usually using a min-heap) to store each node and its distance.

[0147] 2. Algorithm execution:

[0148] Take the node with the smallest distance from the priority queue and update the distances of its adjacent nodes.

[0149] For the current node And every adjacent node , if through arrive The distance is less than the current record Distance, update distance, and Add to the priority queue. The formula is:

[0150]

[0151] in From the starting point to the node The shortest distance, Is a node arrive The distance to the edge.

[0152] 3. Iterative update:

[0153] Repeat the above steps until the priority queue is empty, at which point the shortest paths to all reachable nodes have been found.

[0154] 4. Result extraction:

[0155] Finally, the shortest distance from the starting point to each node will be stored in the distance array middle.

[0156] Through the Dijkstra algorithm, the scheduling system can efficiently plan the shortest path from any designated starting point to all key points in the construction area, ensuring the optimal efficiency of resource allocation and personnel mobilization.

[0157] 4. Execution and feedback module

[0158] In the intelligent scheduling system, the execution and feedback module plays a vital role. It ensures the accurate execution of the scheduling instructions and provides real-time feedback on the execution status so as to make real-time adjustments and optimizations. The module first converts the scheduling plan received from the scheduling decision module into specific operation instructions and sends them to the personnel and mechanical equipment on site through an appropriate communication system. These instructions include but are not limited to movement paths, operation sequences and necessary safety warnings. In order to ensure that the tasks are executed as planned, the module is equipped with an efficient monitoring system, which uses sensors and GPS installed on the equipment to track the location, speed and working status of personnel and equipment in real time. This information is transmitted back to the control center in real time to monitor the progress and quality of task execution. In addition, the execution and feedback module is also responsible for collecting data during the execution process and feeding this data back to the data processing module. The data processing module analyzes this data, evaluates the deviation between the actual execution and the predetermined plan, and identifies possible problems. Based on these analyses, the scheduling decision module can adjust the plan, optimize resource allocation and scheduling strategies to cope with similar situations that may occur in the future.

[0159] 5. User interface module

[0160] In the intelligent scheduling system, the user interface module acts as the core platform for information integration and interactive operation. It is closely connected with the data acquisition module, data processing module, scheduling decision module, and execution and feedback module in the system to ensure smooth transmission and effective management of information.

[0161] The user interface module directly displays the real-time data collected by the data acquisition module, such as the location information of personnel and equipment, the operating status, etc. These real-time data are analyzed and sorted by the data processing module and converted into information that can be directly used by users, such as risk assessment results and behavior pattern analysis. This information is intuitively displayed in the form of charts, indicators and maps on the interface, so that the dispatcher can quickly understand the on-site situation. In addition, the dispatch plan and instructions generated by the dispatch decision module are also transmitted to the operator through the user interface. The interface provides operation functions such as task allocation, path selection and priority adjustment. The dispatcher can make decisions and adjustments based on real-time data and system recommendations. The execution results and status feedback collected by the execution and feedback module are also updated in real time on the user interface, helping the dispatcher to monitor the progress and effect of the task and make necessary adjustments in time. The user interface has become a bridge for information exchange between modules, which not only enhances the responsiveness and adaptability of the dispatch system, but also greatly improves the convenience of operation and the overall efficiency of the system.

[0162] Compared with the prior art, the intelligent dispatching system of the present invention has the following significant technical advantages and beneficial effects:

[0163] 1. Real-time data integration and optimized decision-making: Through highly integrated data collection, processing and decision-making modules, this system can respond to changes on the construction site in real time and optimize scheduling decisions, thereby significantly improving construction efficiency and resource utilization.

[0164] 2. Dynamic adjustment and feedback mechanism: The real-time monitoring and dynamic adjustment functions of the execution and feedback modules enable the scheduling system to respond to on-site emergencies immediately, reduce project delays and cost overruns, and ensure construction progress and quality.

[0165] 3. Enhanced user interaction experience: The user interface module provides an intuitive and easy-to-operate interactive platform, enabling dispatchers to effectively monitor and manage the entire construction process, improving operational convenience and decision-making accuracy.

[0166] 4. Risk management and safety assurance: The system’s risk assessment and management functions enhance construction safety, significantly reduce the accident rate and ensure the safety of personnel and equipment by predicting and preventing potential risks.

[0167] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent scheduling system for tunnel construction operations, characterized in that: The construction system includes: a data acquisition module, a data processing module, a scheduling decision module, an execution and feedback module and a user interface module; wherein, The data acquisition module is used to capture the precise location and operating status of personnel and vehicles on the tunnel construction site in real time, and to format the collected data into standard longitude and latitude coordinates and specific status descriptions, and transmit them to the data processing module in real time for storage and backup; The data processing module is used to conduct in-depth analysis and processing of a large amount of raw data received from the data acquisition module, including data cleaning and preprocessing, feature extraction, behavior analysis and risk assessment, and send the analyzed and processed data to the scheduling decision module to ensure that the scheduling decision module can respond based on accurate and real-time information; The scheduling decision module is used to convert the collected and processed data into an actual executable scheduling plan, and send the scheduling plan to the execution and feedback module; including task priority allocation, resource allocation and path planning; The execution and feedback module is used to convert the scheduling plan received from the scheduling decision module into specific operation instructions, and send them to the personnel and mechanical equipment on site through the communication system; The user interface module is used to directly display the real-time data collected by the data acquisition module. These real-time data are analyzed and organized by the data processing module and converted into information that can be directly used by users. This information is intuitively displayed in the form of charts, indicators and maps on the interface, so that dispatchers can quickly understand the on-site situation; the scheduling plans and instructions generated by the scheduling decision module are also transmitted to the operators through the user interface. The interface provides operation functions, and the dispatchers make decisions and adjustments based on real-time data and system recommendations; the execution results and status feedback collected by the execution and feedback module are also updated in real time on the user interface, helping dispatchers monitor the progress and effects of tasks and make necessary adjustments in a timely manner.

2. The intelligent scheduling system for tunnel construction operations according to claim 1, characterized in that: The process of data cleaning and preprocessing is as follows: Remove noise and outliers: First, use the Z-Score method to standardize the data; for each data point, calculate its deviation from the mean and divide it by the standard deviation. The resulting Z value represents the standard score of the data point; the formula is: ； in is the original data point, is the average value, is the standard deviation; when When the absolute value of is greater than 3, the data point is considered an outlier and is removed; Data smoothing: Kalman filtering algorithm is applied to smooth the position data to eliminate random fluctuations caused by equipment errors or environmental factors. Kalman filtering optimizes the accuracy of data through two steps: prediction and update. Its basic formula is: ； ； in is the current estimated state, is the actual measured value, is the observation matrix, is the Kalman gain, is the measurement noise covariance matrix, is the forecast error covariance.

3. The intelligent scheduling system for tunnel construction operations according to claim 1, characterized in that: The feature extraction process is as follows: Feature extraction is to extract useful information from preprocessed data to provide input for subsequent behavior analysis and risk assessment; in the tunnel construction scheduling system, key features include speed, acceleration and dwell time; Velocity and acceleration calculations: speed It is calculated directly from the location information provided by GPS according to the time change; the formula is: ； in, is the position change, is the time variation; Acceleration is the time derivative of velocity, calculated from two consecutive velocity measurements: ； in, is the velocity change; Duration: For a specific area, the residence time refers to the total time a vehicle or person spends in the area; by analyzing the location time series data, when the timestamps of entering and leaving a specific area are recorded, the residence time is the difference between these two timestamps.

4. The intelligent scheduling system for tunnel construction operations according to claim 1, characterized in that: The process of the behavioral analysis is as follows: Behavioral analysis using SVM: Leverage extracted data including vehicle speed, acceleration, and dwell time in specific areas; define the types of behavior that need to be identified, including idling, moving, or loading activities; label the data accordingly using manual input from historical observations or expert insights; Model training: Use labeled data sets to train SVM; SVM will learn to distinguish different behaviors based on input features; including selecting kernel type and adjusting parameters, regularization parameters and kernel coefficients, the formula is as follows: ； in, is the input feature, is the category label, are the learned parameters, is the kernel function, is the bias term; Cross-validation: Use k-fold cross-validation technique to evaluate the performance of SVM to ensure that the model has good generalization ability to unseen data; adjust parameters based on accuracy, precision, recall and F1 score; Behavior prediction: Once training and validation are complete, deploy the SVM model to predict the behavior of new data in real time. The model will classify vehicle and human activities based on current speed, acceleration, and other contextual data. Output: The final output of the SVM analysis will be the classification of each observation instance into the predefined behavioral categories.

5. The intelligent scheduling system for tunnel construction operations according to claim 1, characterized in that: The risk assessment process is as follows: Risk assessment is to predict potential security risks based on the extracted features and behavior analysis results. It includes the following steps: Step 1: Define risk indicators: First determine which behaviors or conditions are high risk; Step 2: Collect Data: Collect corresponding data from feature extraction and behavior analysis modules based on defined risk indicators; Step 3: Risk Calculation: A logistic regression model is applied to assess risk; the model takes the characteristics of various behaviors and states as input and outputs a risk probability value, calculated as: ； in, is the probability of the risk occurring. are input features, are model parameters; Step 4: Risk level classification: Classify behaviors into different risk levels based on risk probability; and set thresholds to distinguish these levels; Step 5: Report and Respond: The risk assessment results are reported to the dispatch center, and corresponding prevention or intervention measures are taken according to the risk level; Based on the extracted features and behavior analysis results, a logistic regression model is used to evaluate potential security risks; the model calculates the risk probability using the following formula: ； in is the probability of the event occurring, are model parameters, It is an influencing factor; Through the above comprehensive data processing strategy, the intelligent scheduling system can effectively extract valuable information from huge data sets and provide scientific and accurate support for scheduling decisions.

6. The intelligent scheduling system for tunnel construction operations according to claim 1, characterized in that: The specific process of the path planning is as follows: The Dijkstra algorithm is used for path planning, as follows: Step 1: Initialization: Set the distance of all nodes to infinity and the distance of the starting point to 0; Create a priority queue to store each node and its distance; Step 2: Algorithm Execution: Take the node with the smallest distance from the priority queue and update the distance of its adjacent nodes; For the current node And every adjacent node , if through arrive The distance is less than the current record Distance, update distance, and Add to the priority queue; the formula is: ； in From the starting point to the node The shortest distance, Is a node arrive The distance of the edge; Step 3: Iterative Update: Repeat the above steps until the priority queue is empty, at which point the shortest paths to all reachable nodes have been found; Step 4: Result extraction: Finally, the shortest distance from the starting point to each node will be stored in the distance array middle; Through the Dijkstra algorithm, the scheduling system can efficiently plan the shortest path from any designated starting point to all key points in the construction area, ensuring the optimal efficiency of resource allocation and personnel mobilization.

Images