A construction vehicle optimized scheduling method suitable for drill and blast underground construction

Abstract

The application provides a construction vehicle optimization scheduling method suitable for drill and blast underground construction, relates to the technical field of underground construction, and comprises the following steps: firstly, engineering basic data is acquired and basic scheduling parameters of single blasting are calculated, which are input into a construction scheduling model, and a vehicle scheduling instruction set is output by solving the model through an optimization algorithm; in the construction, the construction and vehicle operation states are monitored in real time, a re-scheduling mechanism is triggered when a dynamic scheduling event is monitored, model input parameters are updated and an updated instruction set is generated by re-solving, and the construction vehicle is controlled to continuously work; the method is suitable for the case that the space is limited, the construction section is multiple, and the vehicle resources are limited, can respond to dynamic changes in real time, reduce the idle rate of the vehicle, and improve the construction efficiency; the total scheduling construction period, cost and construction safety are taken as comprehensive optimization targets, the collaborative work of multiple types of vehicles in a complex environment is ensured, and the demand of the drill and blast construction process complexity and the dynamic change of the work environment is met.

Description

Technical Field

[0001] This application generally relates to the field of underground construction technology, and specifically to an optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method. Background Technology

[0002] Drill-and-blast method, a common approach for constructing underground tunnels, caverns, and mine roadways, involves a series of procedures including drilling, charging, blasting, ventilation, muck removal, and support. The construction process is complex, and the working environment is dynamically changing, requiring the coordinated operation of various construction vehicles. However, in traditional construction, the coordination between these procedures relies primarily on manual experience, lacking systematic coordination and dynamic adjustment capabilities. Due to the limited space in underground roadways, numerous construction sections, and limited vehicle resources, existing scheduling methods that rely on experience or static planning are unable to respond in real-time to dynamic changes such as path conflicts and procedure delays, resulting in high vehicle idle rates, long muck removal times, and overall low construction efficiency. Summary of the Invention

[0003] In view of the above-mentioned defects or deficiencies in the existing technology, it is desirable to provide an optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method that can solve the above-mentioned technical problems.

[0004] This application provides an optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method, comprising the following steps:

[0005] Acquire basic engineering data and calculate the basic scheduling parameters for a single blasting operation based on the basic engineering data;

[0006] The basic scheduling parameters are input into the construction scheduling model to generate an initial vehicle scheduling instruction set; wherein, the construction scheduling model is configured to: apply preset scheduling rules, and use the total scheduling period, scheduling cost and construction safety as comprehensive optimization objectives, and solve them through an optimization algorithm;

[0007] The construction vehicles are controlled to perform operations according to the initial vehicle dispatch instruction set, and the construction status and vehicle operation status are monitored in real time during the construction process.

[0008] When a dynamic scheduling event is detected, a rescheduling mechanism is triggered. Based on the current construction status and vehicle operation status, the input parameters of the construction scheduling model are updated, and the construction scheduling model is re-solved to generate an updated vehicle scheduling instruction set to control the construction vehicles to continue to perform operations.

[0009] According to the technical solution provided in this application, the basic engineering data includes at least: tunnel cross-sectional parameters, blasting design parameters, vehicle operation capacity parameters, and process time parameters;

[0010] Acquire basic engineering data, and calculate the basic scheduling parameters for a single blasting operation based on the basic engineering data, specifically including the following steps:

[0011] Based on the tunnel cross-sectional parameters and the blasting design parameters, the amount of rock blasted in a single blast is calculated.

[0012] Based on the parameters of the amount of blasted rock and the vehicle's operational capacity, the required number of transport trucks and the time required to load and unload the waste rock are calculated.

[0013] Based on the time for loading waste rock and the process time parameters, the total fixed process time is calculated.

[0014] According to the technical solution provided in this application, the basic scheduling parameters are input into the construction scheduling model to generate an initial vehicle scheduling instruction set; wherein, the construction scheduling model is configured to: apply preset scheduling rules, and use the total scheduling period, scheduling cost, and construction safety as comprehensive optimization objectives, and solve them through an optimization algorithm, specifically including the following steps:

[0015] Based on the basic scheduling parameters, the preset scheduling rules are applied to generate vehicle scheduling logic; the preset scheduling rules include: vehicle process matching rules and vehicle path priority rules.

[0016] Taking the overall scheduling period, scheduling cost, and construction safety as comprehensive optimization objectives, the vehicle scheduling logic is solved through an optimization algorithm, and the globally optimized initial vehicle scheduling instruction set is output.

[0017] According to the technical solution provided in this application, before generating vehicle scheduling logic based on the basic scheduling parameters and the preset scheduling rules, the preset scheduling rules include vehicle process matching rules and vehicle path priority rules, and further include the following steps:

[0018] Construct a spatial topology network for the tunnels and number the intersections, construction sections, vehicle types, and process statuses within the tunnels.

[0019] According to the technical solution provided in this application, the vehicle process matching rule is implemented through a destination function. The input of the destination function is the process number and the construction section number, and the output of the destination function is the type of vehicle to be dispatched for the corresponding process.

[0020] According to the technical solution provided in this application, the vehicle route priority rule is implemented through a dynamic priority function, which assigns priority based on the vehicle's load status, with loaded vehicles having a higher priority than unloaded vehicles.

[0021] According to the technical solution provided in this application, taking the total scheduling period, scheduling cost, and construction safety as comprehensive optimization objectives, the vehicle scheduling logic is solved through an optimization algorithm to output the globally optimized initial vehicle scheduling instruction set, specifically including the following steps:

[0022] A multi-objective fitness function is constructed to comprehensively evaluate any vehicle scheduling scheme; the multi-objective fitness function is a weighted sum based on the total scheduling time function, scheduling cost function, and safety score function;

[0023] Using the multi-objective fitness function as the evaluation criterion, the optimization algorithm is run to iteratively search and compare various vehicle scheduling schemes generated based on the vehicle scheduling logic, and the vehicle scheduling scheme with the optimal fitness value is output as the initial vehicle scheduling instruction set.

[0024] According to the technical solution provided in this application, using the multi-objective fitness function as the evaluation criterion, the optimization algorithm is run to iteratively search and compare various vehicle scheduling schemes generated based on the vehicle scheduling logic, and the vehicle scheduling scheme that optimizes the multi-objective fitness function value is output as the initial vehicle scheduling instruction set. Specifically, the following steps are included:

[0025] S1: Encode different vehicle scheduling schemes to generate an initial population containing multiple vehicle scheduling schemes;

[0026] S2: Using the multi-objective fitness function as the evaluation criterion, repeatedly perform selection, crossover, and mutation operations on the initial population to generate the next generation population;

[0027] S3: Each time, use the generated next generation population as a new initial population and repeat step S2 until the iteration termination condition is met.

[0028] S4: Decode the individual with the best fitness value in the population obtained when the termination condition is met, and output it as the initial vehicle scheduling instruction set.

[0029] According to the technical solution provided in this application, when a dynamic scheduling event is detected, a rescheduling mechanism is triggered. Based on the current construction status and vehicle operation status, the input parameters of the construction scheduling model are updated, and the construction scheduling model is re-solved to generate an updated vehicle scheduling instruction set to control the construction vehicles to continue performing operations. Specifically, the steps include the following:

[0030] By comparing the real-time construction status with the vehicle operation status and the currently executed vehicle dispatch instruction set, it is determined whether a dynamic dispatch event has occurred; the dynamic dispatch event includes the vehicle staying at a critical node for an extended period of time, the actual time consumed by the process exceeding the planned threshold, or the vehicle reporting a fault.

[0031] When the dynamic scheduling event is determined to occur, the rescheduling mechanism is triggered; based on the current construction status and vehicle operation status, the input parameters of the construction scheduling model are updated, and the input parameters include at least: the location of the affected vehicles, the status of the obstructed process, and the available equipment resources;

[0032] The input parameters are input into the construction scheduling model to control the optimization algorithm to solve the problem again and generate an updated vehicle scheduling instruction set.

[0033] The updated vehicle dispatch instruction set is sent to the corresponding construction vehicles to control the vehicles to continue performing operations.

[0034] The beneficial effects of this application are as follows:

[0035] This application provides an optimized scheduling method for construction vehicles applicable to drill-and-blast underground construction. It first acquires basic engineering data and calculates the basic scheduling parameters for a single blast, then inputs these parameters into a construction scheduling model. An optimization algorithm solves for and outputs a vehicle scheduling instruction set. Simultaneously, the construction status and vehicle operation status are monitored in real time during construction. When a dynamic scheduling event is detected, a rescheduling mechanism is triggered. Based on the current status, the input parameters of the construction scheduling model are updated, and a new vehicle scheduling instruction set is generated to control the continuous operation of construction vehicles. This effectively solves the problems of traditional drill-and-blast underground construction, which relies on manual experience to arrange work sequence connections and lacks systematic coordination and dynamic adjustment capabilities. Addressing the realities of limited underground tunnel space, numerous construction sections, and limited vehicle resources, this method overcomes the limitations of static planning, enabling real-time response to dynamic changes such as path conflicts and work sequence delays. It reduces the idle rate of construction vehicles, shortens muck removal time, and significantly improves overall construction efficiency. Meanwhile, its design, which takes the overall scheduling period, scheduling cost and construction safety as comprehensive optimization goals, has achieved coordinated optimization of schedule, cost and safety on the basis of ensuring the orderly progress of construction and improving operation efficiency. It has ensured the efficient collaborative operation of multiple types of construction vehicles in complex construction environments, enhanced the scientificity and flexibility of drilling and blasting underground construction scheduling, and better adapted to the needs of complex drilling and blasting construction processes and dynamic and ever-changing operating environments. Attached Figure Description

[0036] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0037] Figure 1 This is a flowchart of an optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method, provided in Embodiment 1 of this application;

[0038] Figure 2 This is a schematic diagram of the tunnel cross-section provided in Embodiment 1 of this application. Detailed Implementation

[0039] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0040] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0041] Example 1

[0042] Please refer to Figure 1 This application provides a method for optimizing the scheduling of construction vehicles applicable to underground construction using the drill-and-blast method, comprising the following steps:

[0043] S100: Obtain basic engineering data and calculate the basic scheduling parameters for a single blasting operation based on the basic engineering data;

[0044] Optionally, the basic engineering data should include at least: tunnel cross-section parameters, blasting design parameters, vehicle operation capacity parameters, and process time parameters;

[0045] Specifically, such as Figure 2 As shown, the tunnel cross-sectional parameters mainly refer to the aggregate dimensions of the tunnel, which are used to characterize the cross-sectional size of the construction space, including the tunnel bottom width B and the tunnel cross-sectional height h.

[0046] Blasting design parameters mainly refer to the advance length L of a single blast, which determines the distance the roadway advances in each work cycle;

[0047] Vehicle operation capability parameters include the performance indicators of key construction equipment, including the rated load of transport trucks (m1), the bucket load of scrapers (m2), and the loading and unloading time of a single bucket of scrapers (t1).

[0048] Process time parameters mainly refer to the time consumed by each fixed link determined according to process requirements and historical data, including: drilling and charging time T1, ventilation time T2, and support time T4.

[0049] After obtaining the aforementioned basic engineering data, the system performs a series of coherent calculations to derive the basic scheduling parameters used for subsequent intelligent scheduling decisions.

[0050] Step S100 specifically includes the following steps:

[0051] S110: The amount of rock blasted in a single blast is calculated based on the tunnel cross-section parameters and blasting design parameters.

[0052] S120: Based on the blasting rock volume and vehicle operation capacity parameters, calculate the required number of transport trucks and the time required to load waste rock;

[0053] S130: Calculate the total fixed process time based on the time for loading waste rock and the process time parameters.

[0054] Specifically, in this embodiment, the tunnel cross-section is usually a three-center arch cross-section. First, based on the tunnel bottom width B and the tunnel cross-section height h, the cross-sectional area S is calculated using Formula 1.

[0055] Formula 1;

[0056] Subsequently, based on the cross-sectional area S, the advance length L of a single blast, and the rock density ρ, the amount of rock blasted in a single blast is calculated using Formula 2; the amount of rock blasted is the fundamental indicator for measuring the total amount of slag removal in this cycle.

[0057] Formula 2;

[0058] Subsequently, based on the amount of blasted rock M and the rated load m1 of the transport trucks, the number of transport trucks n required to theoretically complete the transportation of all waste rock was calculated using Formula 3; the required number of transport trucks provides a direct quantitative basis for assessing the demand for transportation resources and for vehicle allocation.

[0059] Formula 3;

[0060] Subsequently, based on the amount of blasted rock M, the load capacity of the loader bucket m2, and the loading and unloading time of a single loader bucket t1, the loading and unloading time of waste rock T3 was calculated using Formula 4.

[0061] Formula 4;

[0062] Finally, the drilling and charging time T1, ventilation time T2, support time T4, and waste rock loading time T3 are summed to obtain the total fixed process time T, i.e. This timeframe serves as the baseline timeframe for subsequent optimization of vehicle scheduling and reduction of non-productive waiting time.

[0063] Through a series of structured data acquisition and calculations, this application integrates scattered, multi-dimensional raw engineering information into a set of key, quantifiable basic scheduling parameters. These basic scheduling parameters clearly define workload, transportation capacity, loading time, and total construction period, providing indispensable data support for the construction scheduling model to conduct scientific resource planning, route planning, and dynamic decision-making.

[0064] S200: Input the basic scheduling parameters into the construction scheduling model to generate an initial vehicle scheduling instruction set; wherein, the construction scheduling model is configured to: apply preset scheduling rules, and use the total scheduling period, scheduling cost and construction safety as comprehensive optimization objectives, and solve them through optimization algorithms;

[0065] Specifically, the construction scheduling model is a comprehensive computing system that integrates environmental modeling, rule engine and optimization algorithm.

[0066] Optionally, step S200 specifically includes the following steps:

[0067] S210: Based on basic scheduling parameters, apply preset scheduling rules to generate vehicle scheduling logic; preset scheduling rules include: vehicle process matching rules and vehicle path priority rules.

[0068] S220: Taking the overall scheduling period, scheduling cost and construction safety as comprehensive optimization objectives, it solves the vehicle scheduling logic through optimization algorithms and outputs a globally optimized initial vehicle scheduling instruction set.

[0069] It should be noted that this application transforms basic engineering parameters into structured vehicle scheduling logic through preset scheduling rules. Based on this, it drives a multi-objective optimization algorithm for global optimization, and finally automatically generates an initial vehicle scheduling instruction set that achieves the optimal balance between total scheduling period, scheduling cost and construction safety. This realizes the intelligent leap from static data to dynamic optimization decision-making, and lays the core decision-making foundation for efficient, economical and safe construction execution.

[0070] Furthermore, prior to step S210, the following steps are also included:

[0071] S201: Construct a spatial topology network for the tunnels and number the intersections, construction sections, vehicle types, and process statuses within the tunnels.

[0072] Specifically, the first step is to construct a lane space topology network to uniquely number key elements, thereby forming a precise scheduling language:

[0073] The key intersections within the roadway where vehicles interact or make changes in their routes will be numbered sequentially as A0, A1, A2, ..., A1. i The construction sections requiring drilling, slag removal, and support operations shall be numbered sequentially as 1, 2, 3, ..., a according to the construction sequence or location.

[0074] Each type of construction vehicle participating in the collaborative operation is assigned a unique type identifier to accurately specify the object in the dispatching instructions; in this embodiment, the rock drilling rig is numbered Z, the loader is numbered C, and the transport truck is numbered Y.

[0075] The continuous drilling and blasting construction cycle is decomposed into discrete, sequential logical stages, and a unique process number is defined for each stage to represent the current operation status. In this embodiment, the following are defined: 0 is drilling and charging, 1 is blasting and ventilation, 2 is shoveling and loading waste rock, 3 is transporting and removing slag, and 4 is drilling and support.

[0076] Specifically, the core of step S210 is to use preset scheduling rules to transform construction requirements into a set of clear vehicle action instructions and constraints, namely vehicle scheduling logic. The generation of vehicle scheduling logic mainly depends on the application of vehicle process matching rules and vehicle path priority rules.

[0077] The vehicle process matching rules are implemented through a destination function. The input of the destination function is the process number and the construction section number, and the output of the destination function is the type of vehicle that should be dispatched for the corresponding process.

[0078] Specifically, in this embodiment, the vehicle process matching rule is implemented through a destination function, which is defined as follows: The input to the destination function comes from a pre-defined coding system, where x is the process number, a discrete variable used to uniquely identify a specific stage in the drill-and-blast construction cycle. For example, x=0 represents "drilling and charging", x=1 represents "blasting and ventilation", x=2 represents "shoveling and loading waste rock", x=3 represents "transporting and removing slag", and x=4 represents "drilling and support". a is the construction section number, used to uniquely specify the location of the specific working face where the above process needs to be performed, such as construction section 1, construction section 2, etc.

[0079] For any given process number x and construction section number a, the destination function outputs the vehicle type number that should be dispatched for the corresponding process;

[0080] In this embodiment, the mapping relationship is specifically configured as follows: This means that when the process is "drilling and charging" (x=0), the function outputs the vehicle type number Z, and the system automatically generates a dispatch instruction to dispatch the rock drilling rig to construction section a.

[0081] This means that when the process is "shoveling and loading waste rock" (x=2), the function outputs vehicle type number C, and the system automatically generates a dispatch instruction to dispatch the shovel to construction section a.

[0082] This means that when the process is "transporting slag" (x=3), the function outputs vehicle type number Y, and the system automatically generates a dispatch instruction to dispatch a transport truck to construction section a.

[0083] Through destination functions, the abstract process requirements in the construction plan are translated into specific and executable vehicle scheduling logic in real time and automatically. This completely replaces the traditional model that relies on the personal experience of dispatchers for manual judgment and assignment, fundamentally ensuring the absolute accuracy, consistency and efficiency of task distribution, effectively avoiding process delays or resource conflicts caused by vehicle dispatch errors, and laying the foundation for intelligent decision-making of the entire scheduling system.

[0084] The vehicle route priority rule is implemented through a dynamic priority function, which assigns priority based on the vehicle's load status, with loaded vehicles having a higher priority than unloaded vehicles.

[0085] Specifically, the vehicle path priority rule is implemented through a dynamic priority function (denoted as δ). The dynamic priority function is the core decision-making logic of the construction scheduling model to resolve vehicle traffic conflicts in the limited space of the tunnel and ensure smooth and safe operations. The design follows the fundamental principles of prioritizing heavy vehicles and safety, and arbitrates and plans the traffic order of vehicles in key areas such as intersections and one-way sections by dynamically calculating a priority value for each vehicle.

[0086] In the blasting and decommissioning process, fully loaded trucks carry the blasted waste rock for each cycle, and their transportation efficiency directly affects the overall cycle's duration. Therefore, the dynamic priority function's allocation logic clearly stipulates that loaded (fully loaded) vehicles have higher priority than empty vehicles, ensuring the passage of loaded vehicles first, minimizing delays on the critical path, and aligning with lean construction management objectives. Simultaneously, since the energy consumption and mechanical losses from starting, stopping, and accelerating loaded vehicles are significantly higher than those from empty vehicles, this rule also helps reduce overall operating costs.

[0087] In this embodiment, the basic assignment example of the dynamic priority function δ is as follows:

[0088] Unloaded transport truck: δ=1;

[0089] Fully loaded transport truck: δ=100;

[0090] Rock drilling rig or loader: δ=10;

[0091] Furthermore, to address special working conditions and enhance safety, the rule includes extended logic: when a vehicle needs to travel in the opposite direction within a single-lane section due to task requirements (such as the removal of a disabled vehicle or the entry of special equipment), its priority will be automatically increased, i.e., doubled on the original assignment; this is equivalent to sending a high-priority warning signal to the system, enabling the construction scheduling model to proactively reserve safety space or arrange avoidance for other vehicles when planning their routes, thereby greatly reducing the risk of oncoming conflicts.

[0092] By employing a dynamic priority function, the construction scheduling model transforms the difficult-to-quantify decision-making process of passing weights into calculable and comparable values ​​based on clearly defined rules. When generating scheduling plans, the model ensures that path delays for high-priority vehicles are minimized. When potential path conflicts are detected in real-time, the system can also quickly arbitrate based on this function value, instructing low-priority vehicles to wait at a safe location or give way. This replaces the uncertainty and safety hazards inherent in traditional models that rely on individual driver judgment or ground personnel coordination, achieving rule-based, orderly, and intelligent management of underground vehicle traffic, providing crucial guarantees for overall construction efficiency and safety.

[0093] Optionally, step S220 specifically includes the following steps:

[0094] S221: Construct a multi-objective fitness function to comprehensively evaluate any vehicle scheduling scheme; the multi-objective fitness function is a weighted sum based on the total scheduling time function, scheduling cost function, and safety score function;

[0095] S222: Using a multi-objective fitness function as the evaluation criterion, an optimization algorithm is run to iteratively search and compare various vehicle scheduling schemes generated based on vehicle scheduling logic, and the vehicle scheduling scheme with the optimal fitness value is output as the initial vehicle scheduling instruction set.

[0096] Specifically, this application constructs a multi-objective fitness function (weighted sum) that integrates total scheduling period, scheduling cost, and construction safety, establishing a unified and quantitative mathematical standard for evaluating the merits of any scheduling scheme. Furthermore, using this function as the sole evaluation criterion and evolutionary guide, the optimization algorithm is driven to systematically iteratively search and compare all feasible schemes defined by the vehicle scheduling logic. The ultimate effect of this process is to automatically eliminate a massive number of local or suboptimal solutions, intelligently identify and output a vehicle scheduling scheme that achieves the optimal fitness function value globally, thereby generating an initial vehicle scheduling instruction set that achieves the best balance among efficiency, economy, and safety.

[0097] Furthermore, the construction of the multi-objective fitness function is achieved through the following steps and formulas:

[0098] First, the sum of total idle time and process waiting time is calculated using Formula 5. This time reflects the non-operational time caused by path conflicts and poor task connection in the scheduling scheme.

[0099] Formula 5;

[0100] Where T5 is the sum of total idle time and process waiting time. Let j be the waiting time for vehicle number j at the passing junction. denoted as j, represents the scheduling waiting time for vehicle j within the task interval, and b represents the number of construction vehicle groups (each group typically includes 1 rock drilling rig, 1 loader, and 2 transport trucks).

[0101] Subsequently, the total scheduling time function is calculated using Formula 6. This function represents the estimated total time required to complete a full construction cycle and is the core indicator for measuring the efficiency of the project schedule.

[0102] Formula Six;

[0103] in: For the overall scheduling time function, The total time for a fixed process.

[0104] Finally, based on the above results, the multi-objective fitness function is obtained through Formulas 7 and 8. This function is the final mathematical standard for comprehensive optimization decision-making.

[0105] Formula 7;

[0106] Formula 8;

[0107] Where ω1, ω2, and ω3 are weighting coefficients (which can be flexibly adjusted according to the project stage to reflect different management focuses; for example, in the early stage of tunneling, ω1 is 0.6, ω2 is 0.3, and ω3 is 0.1); Cost is the scheduling cost function. This is the fuel consumption cost coefficient per unit distance traveled. The equipment depreciation cost factor per unit waiting time. Let j be the distance traveled by vehicle number j. The waiting time for vehicle number j at the passing junction;

[0108] Safety_Score is a safety scoring function used to assess the safety risk level of a dispatching plan. The value of the safety scoring function is negatively correlated with the meeting risk of the dispatching plan. Meeting risk is divided into two levels based on the comparison between the expected number of meeting trips and the total number of dispatched vehicles: when the expected number of meeting trips is less than or equal to twice the total number of dispatched vehicles (2 * total number of dispatched vehicles), it is considered a low-risk state; while when the expected number of meeting trips is greater than twice the total number of dispatched vehicles (2 * total number of dispatched vehicles), it is considered a medium-to-high-risk state. In the low-risk state, the meeting risk... In medium- to high-risk conditions, the risk of vehicles meeting each other is high. Subsequently, the safety score function is calculated according to Formula 9. The estimated number of meeting points is determined based on the task path data of the specific plan, representing the total number of points where all vehicles need to coordinate meeting points during the scheduling process. This calculation mechanism ensures that the more points requiring meeting point coordination, the lower the safety score the plan receives, thus directly reflecting the safety performance of the scheduling plan in avoiding traffic conflicts.

[0109] Formula Nine.

[0110] Optionally, step S222 specifically includes the following steps:

[0111] S1: Encode different vehicle scheduling schemes to generate an initial population containing multiple vehicle scheduling schemes;

[0112] S2: Using a multi-objective fitness function as the evaluation criterion, the initial population is repeatedly subjected to selection, crossover, and mutation operations to generate the next generation population;

[0113] S3: Each time, use the generated next generation population as a new initial population and repeat step S2 until the iteration termination condition is met.

[0114] S4: Decode the individual with the best fitness value in the population obtained when the termination condition is met, and output it as the initial vehicle scheduling instruction set.

[0115] Specifically, this application utilizes a genetic algorithm to achieve global optimization search. This algorithm mimics the mechanism of natural evolution, performing an intelligent global search within the solution space defined by the vehicle scheduling logic. The entire process begins with encoding the vehicle scheduling scheme and initializing the population: using matrix encoding, a... A chromosome matrix of dimension b is used, where b is the number of construction vehicle groups (each group typically includes 1 rock drill rig, 1 loader, and 2 transport trucks), i is the total number of critical intersections in the roadway where vehicles interact or make changes in their paths, and v is the total number of construction sections. In the chromosome matrix, each row corresponds to a complete scheduling instruction sequence for one vehicle, and each gene position corresponds to a node (intersection or construction section) in the roadway network, with values ​​from the set {0, 1, 2}, representing "do not pass through this node", "wait at this node", and "pass directly through this node", respectively. By randomly generating a large number of such chromosomes, an initial population containing diverse feasible vehicle scheduling schemes is formed.

[0116] Subsequently, the algorithm enters an iterative evolutionary loop using a multi-objective fitness function F as the sole evaluation criterion. In each generation, selection, crossover, and mutation operations are performed sequentially: First, selection is performed using a tournament selection strategy, randomly selecting several individuals from the initial population to form a competition group, retaining the individuals with the highest fitness to maintain population diversity and transmit superior genes. Next, crossover is performed using a two-point crossover strategy, randomly selecting two cut points and using a preset crossover probability (e.g., ...). The chromosome segments of paired parent individuals are exchanged to generate new offspring that incorporate traits of both parents. Finally, a mutation operation is performed with a low probability (e.g., ...). The gene loci of an individual are randomly flipped, and a new value is randomly selected from its set of values ​​to replace the original value, in order to introduce new characteristics and help the algorithm escape local optima. After generating a new generation of population through the above operations, it is used as the initial population for the next iteration, and this evaluation-selection-crossover-mutation evolutionary process is repeated cyclically.

[0117] The iterative process will continue until a preset termination condition is met, namely, the number of generations reaches the upper limit (e.g., 200 generations) or the fitness value of the best individual fluctuates less than a preset threshold (e.g., 10) over several consecutive generations (e.g., 20 generations). -3 When the algorithm terminates, the individual with the highest fitness value is selected from the final population, and its chromosome is decoded to restore the specific vehicle operation sequence, driving path, and node action instructions. This instruction sequence is the initial vehicle scheduling instruction set obtained through global optimization search, thus achieving the goal of automatically finding the optimal solution under multiple constraints and outputting a comprehensive optimal scheduling scheme.

[0118] S300: Controls construction vehicles to perform operations according to the initial vehicle dispatch instruction set, and monitors the construction status and vehicle operation status in real time during the construction process;

[0119] Specifically, the initial vehicle dispatch instruction set is sent to the onboard intelligent terminals of each construction vehicle via a wireless communication network. This instruction set not only includes the target location but also a detailed spatiotemporal action plan, clearly specifying the driving route of each vehicle within a specific time window, the pre-set actions at key nodes (such as intersections) (such as waiting at the node or passing directly through the node), and the type of task to be performed (such as loading slag or transporting). The vehicle driver or the autonomous driving system executes the operation according to the terminal guidance, enabling the dispersed equipment to operate in a strictly unified and coordinated rhythm, thereby avoiding the coordination gaps and resource conflicts common in traditional manual dispatching.

[0120] S400: When a dynamic scheduling event is detected, a rescheduling mechanism is triggered. Based on the current construction status and vehicle operation status, the input parameters of the construction scheduling model are updated, and the construction scheduling model is re-solved to generate an updated vehicle scheduling instruction set to control the construction vehicles to continue performing operations. The specific steps include the following:

[0121] By comparing the monitored construction status with the vehicle operation status and the currently executed vehicle dispatch instruction set, it is determined whether a dynamic dispatch event has occurred. Dynamic dispatch events include vehicles staying at key nodes for too long, actual process time exceeding the planned threshold, or vehicles reporting malfunctions.

[0122] When a dynamic scheduling event is detected, a rescheduling mechanism is triggered; based on the current construction status and vehicle operation status, the input parameters of the construction scheduling model are updated. The input parameters include at least: the location of the affected vehicles, the status of the obstructed process, and the available equipment resources.

[0123] Input parameters are input into the construction scheduling model, and the control optimization algorithm is re-solved to generate an updated vehicle scheduling instruction set.

[0124] The updated vehicle dispatch instruction set is sent to the corresponding construction vehicles to control the vehicles to continue performing operations.

[0125] Specifically, the system continuously compares real-time data (including precise vehicle location, speed, load, and equipment health status) reported by IoT sensors (such as cameras and radar at key intersections) and vehicle terminals with the currently executing vehicle dispatching command set at millisecond levels. When the system identifies patterns such as "a transport truck stays at the A5 intersection for more than 8 minutes (far exceeding the planned passage time)," "the single-bucket cycle time of the loader at construction section 3 abnormally extends from 5 minutes to 8 minutes," or "a vehicle actively reports an engine failure," it automatically determines that a dynamic dispatching event has occurred. This indicates that the initial static optimization scheme can no longer fully adapt to the actual situation on site, and emergency adjustments must be initiated.

[0126] Once the event is confirmed, the system immediately triggers a rescheduling mechanism. This mechanism is not a simple local fine-tuning, but rather initiates a complete re-optimization process. First, the input parameters of the construction scheduling model are rapidly updated based on the current construction status and vehicle travel status (e.g., the exact location of a disabled vehicle, the progress of a process hindered by a large rock, and the availability of backup equipment resources in the surrounding area). This step is crucial, as it ensures that the re-solved model is based on the latest and most realistic field constraints, rather than outdated theoretical assumptions.

[0127] Following this, the system control optimization algorithm (such as a genetic algorithm) restarts the solution process with the updated parameters as input. To improve response speed and meet the stringent timeliness requirements on-site (e.g., generating a new solution within 30 seconds), the algorithm can automatically switch to a fast response mode. In this mode, the algorithm's search strategy will be adjusted accordingly; for example, the genetic algorithm can temporarily increase the probability of mutation operations (e.g., from the basic 0.02 to 0.05) to explore a wider solution space in a shorter time and accelerate optimization convergence. After the algorithm runs, it outputs a completely new global optimization result that fully considers the latest conflicts and constraints, i.e., the updated vehicle scheduling instruction set.

[0128] Finally, the system accurately distributes the updated vehicle dispatch instruction set to the on-board terminals of the affected vehicles via wireless network. The updated vehicle dispatch instruction set seamlessly replaces the parts of the original plan that have become invalid or no longer applicable. For example, it plans a route back to the repair shop for a disabled vehicle and a detour route for other vehicles, or adds a backup loader for a construction section that is blocked. The vehicles receive and execute the new instructions, thus quickly returning from a disordered or delayed state to a new, collaborative and efficient work path. The entire process from event perception to the issuance of new orders is highly automated and intelligent, transforming the traditional passive response that relies on manual reporting, experience judgment, and lengthy coordination into a proactive closed-loop optimization led by the system and completed within minutes. This greatly improves the resilience of the construction system in the face of uncertainty and the overall operational efficiency.

[0129] Working Principle: This application first acquires basic engineering data and calculates the basic scheduling parameters for a single blast. These parameters are then input into a construction scheduling model, where an optimization algorithm solves for and outputs a vehicle scheduling instruction set. Simultaneously, the construction status and vehicle operation status are monitored in real time during construction. When a dynamic scheduling event is detected, a rescheduling mechanism is triggered. Based on the current status, the input parameters of the construction scheduling model are updated, and a new vehicle scheduling instruction set is generated to control the continuous operation of construction vehicles. This effectively solves the problems of traditional drill-and-blast underground construction, which relies on manual experience to arrange work sequence connections and lacks systematic coordination and dynamic adjustment capabilities. Addressing the realities of limited underground tunnel space, numerous construction sections, and limited vehicle resources, this method overcomes the limitations of static planning, enabling real-time response to dynamic changes such as path conflicts and work sequence delays. This reduces the idle rate of construction vehicles, shortens muck removal time, and significantly improves overall construction efficiency. Meanwhile, its design, which takes the overall scheduling period, scheduling cost and construction safety as comprehensive optimization goals, has achieved coordinated optimization of schedule, cost and safety on the basis of ensuring the orderly progress of construction and improving operation efficiency. It has ensured the efficient collaborative operation of multiple types of construction vehicles in complex construction environments, enhanced the scientificity and flexibility of drilling and blasting underground construction scheduling, and better adapted to the needs of complex drilling and blasting construction processes and dynamic and ever-changing operating environments.

[0130] To facilitate understanding, the optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method in this application will be briefly introduced below with reference to specific embodiments.

[0131] (1) Acquisition of basic engineering data and calculation of basic scheduling parameters:

[0132] The cross-section of the three-center arch tunnel was measured three times using a laser profiler, and the average value was taken to determine the tunnel bottom width B = 5m and the tunnel cross-sectional height h = 5m. Based on Formula 1, the cross-sectional area S = 31.55m² was calculated. 2 .

[0133] By reviewing the engineering blasting design documents, the advance length for a single blast was determined to be L = 1.8m; simultaneously, through laboratory core testing, the rock density was obtained as ρ = 2.2t / m³. 3 According to Formula 2, the amount of rock blasted in a single blast is M = 125t.

[0134] There are 8 transport trucks on site, all with a rated load of m1=15t and no load deviation. According to Formula 3, the required number of transport trucks is n=9. Since there are only 8 main transport trucks on site, 1 spare transport truck will be dispatched to supplement the load and ensure that the slag removal capacity is matched.

[0135] Loading tests were conducted on three loaders to determine the bucket load capacity of the loaders (m2 = 6t) and the single-bucket loading and unloading time (t1 = 1min, including loading, turning, and unloading actions). Based on Formula 4, the theoretical loading time for waste rock (T3 = 21min) was calculated. It should be noted that this time is a theoretical calculation value. In the future, the coordinated operation between the vehicles can be optimized through scheduling algorithms (for example, the scheduling algorithm is based on the operating time of the loaders and dynamically schedules the departure and arrival times of the transport trucks to match the arrival time of the transport trucks with the unloading preparation time of the loaders, thereby optimizing the operation connection and reducing waiting time).

[0136] Through three trial cycles of timing on site, the drilling and charging time T1 was determined to be 15 min and the ventilation time T2 to be 60 min. Based on the rock mass stability report, the support time T4 was determined to be 20 min. Substituting the theoretical waste rock loading time T3 to be 21 min, the total fixed process time T was finally calculated to be 116 min.

[0137] (2) Application of spatial topology construction and construction scheduling model

[0138] Construct a spatial topology network for the tunnel, and number the 8 intersections of the tunnel from the entrance to the depth as A0, A1, A2, A3, ..., A7; and number the 4 construction sections as 1, 2, 3, 4 according to the construction sequence, forming a complete node list {A0, A1, A2, A3, A4, A5, A6, A7, 1, 2, 3, 4}.

[0139] Define process and vehicle codes: Define process codes: Drilling and charging is 0, blasting ventilation is 1, shoveling waste rock is 2, transporting slag is 3, and drilling support is 4. Define vehicle type codes: Rock drilling rig is Z, loader is C, and transport truck is Y.

[0140] The vehicle's location is updated in real time through an underground positioning system (e.g., every 10 seconds). For example, if a rock drilling rig is traveling between construction sections A2 and 1, it is marked as Z(A2,1); if a loader is traveling between construction section 3 and A5, it is marked as C(3,A5).

[0141] When the process enters the "drilling and charging" stage, the destination function N(0,a)=Z is called to dispatch the rock drilling rig to the corresponding construction section a; when the process enters the "shoveling and loading waste rock" stage, N(2,a)=C is called to dispatch the loader to the construction section a, ensuring that the vehicle and the process are precisely matched.

[0142] The scheduling system statistics show the waiting time of vehicles at passing points in the initial cycle. Vehicle dispatching waiting time during task intervals Therefore, according to Formula 5, the sum of total idle time and process waiting time is calculated. The total scheduling time function is calculated based on Formula 6. .

[0143] In this embodiment, ω1=0.6, ω2=0.3, and ω3=0.1 are set. The scheduling cost function Cost=280 and the safety score function Safety_Score=5 (the expected number of meeting points in the scheduling scheme is less than twice the total number of vehicles to be scheduled) are calculated and substituted into Formula 7 to obtain the multi-objective fitness function. , as the optimization benchmark.

[0144] (3) Generation of initial scheduling scheme based on genetic algorithm

[0145] Constructing a chromosome matrix: The number of vehicle groups b=2, with a total of 4b=8 core vehicles (2 rock drilling rigs, 2 loaders, and 4 transport trucks). The total number of intersections and cross-sections i+v=8+4=12, so an 8×12 dimensional chromosome matrix is ​​constructed.

[0146] Each line corresponds to the passing point instruction for one vehicle. For example, the instruction line for rock drilling rig 1 is [0,0,2,1,0,0,2,0,0,1,0,0], where "0" means not passing through the node, "1" means waiting at the node, and "2" means passing through the node directly.

[0147] Selection operation: From the initial 50 groups, 5 groups are randomly selected to form 1 tournament group (10 tournament groups in total). The individual with the highest fitness in each group is retained, and the 10 high-quality individuals directly enter the next generation.

[0148] Crossover operation: Employs a two-point crossover strategy (crossover probability) The gene segments of the parent individuals are randomly selected at the cutting point to be exchanged. For example, the gene segments of parent 1 [0,0,2,1,0,0,2,0,0,1,0,0] are exchanged with the gene segments of parent 2 [2,0,1,0,2,1,0,0,1,0,2,0]. After the crossover, the offspring 1 is [0,0,2,0,2,1,2,0,0,1,0,0] and the offspring 2 is [2,0,1,1,0,0,0,0,1,0,2,0].

[0149] Mutation operation: Randomly perform mutation on the gene positions of all individuals (2% mutation probability), such as flipping "1" (waiting at the node) in a transport truck's instruction line to "2" (passing directly through the node), repairing 3 sets of invalid solutions with path conflicts, and ensuring that the proportion of valid solutions in the population is greater than or equal to 95%.

[0150] When the iteration reaches the 78th generation, the fitness value F fluctuates by less than 10 for 20 consecutive generations. -3 If the termination condition is met, the iteration stops.

[0151] Output the vehicle scheduling scheme corresponding to the optimal chromosome matrix, specifying the travel path of each vehicle (e.g., rock drilling rig 1: A0→A2 (directly pass through this node)→A3 (wait at this node)→No.1 construction section (drilling and charging)→A2 (directly pass through this node)→A0) and task sequence (e.g., 0-15min drilling and charging, 15-60min blasting and ventilation, 60-80min No.1 construction section drilling and support). The scheme is distributed to the on-board terminal of each vehicle through the scheduling system.

[0152] (4) Execution of dynamic event triggering and rescheduling mechanism

[0153] Taking the anomaly that occurred at construction section No. 3 on a certain day as an example, the dynamic adjustment process is explained:

[0154] Data collection: Camera and radar data at intersection A5 showed that transport trucks 6 and 7 stayed at intersection A5 for more than 8 minutes, which is a large deviation from the initial plan of "passing in 2 minutes". At the same time, positioning data of loader 3 showed that its loading interval at construction section 3 increased from 5 minutes to 8 minutes, and the speed sensor showed no abnormalities.

[0155] Event determination: Based on the on-site workers' report (the diameter of the blasted rocks exceeded 1.2m), the incident was determined to be "a path conflict caused by a process delay," triggering the rescheduling mechanism.

[0156] Parameter adjustment: Temporarily increase the mutation probability of the genetic algorithm to 0.05, start 5 generations of fast iteration based on the current optimal solution, and recalculate the destination function N(2,3)=C (only keep 1 loader in construction section 3) and N(3,4)=Y (add 1 transport truck to construction section 4).

[0157] Path and priority adjustment: Mark the intersection A5-A6 section as "occupied" in the lane space topology network, and replan the path for transport truck 6 and transport truck 7 as A5→A3→A6→construction section 4; set the priority of the supporting loader 5 to δ=10 to ensure that it has priority to pass through construction section 3.

[0158] Instructions issued: At 14:08, the dispatch system sent new instructions to the relevant vehicles. Scraper 3 received the instruction to "prioritize small stones", transport trucks 6 and 7 received new route instructions, and scraper 5 received the instruction to "arrive at construction section 3 before 14:20".

[0159] Feedback: At 14:22, loader 5 arrived at construction section 3; at 14:28, transport trucks 6 and 7 completed the first slag removal at construction section 4; at 14:30, the congestion at intersection A5 was relieved; the slag removal time delay was reduced from the predicted 40 minutes to 12 minutes, and the dynamic adjustment was completed.

[0160] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for optimizing the scheduling of construction vehicles applicable to underground construction using the drill-and-blast method, characterized in that, Includes the following steps: Acquire basic engineering data and calculate the basic scheduling parameters for a single blasting operation based on the basic engineering data; The basic scheduling parameters are input into the construction scheduling model to generate an initial vehicle scheduling instruction set; wherein, the construction scheduling model is configured to: apply preset scheduling rules, and use the total scheduling period, scheduling cost and construction safety as comprehensive optimization objectives, and solve them through an optimization algorithm; The construction vehicles are controlled to perform operations according to the initial vehicle dispatch instruction set, and the construction status and vehicle operation status are monitored in real time during the construction process. When a dynamic scheduling event is detected, a rescheduling mechanism is triggered. Based on the current construction status and vehicle operation status, the input parameters of the construction scheduling model are updated, and the construction scheduling model is re-solved to generate an updated vehicle scheduling instruction set to control the construction vehicles to continue to perform operations. The basic engineering data includes at least: tunnel cross-section parameters, blasting design parameters, vehicle operation capacity parameters, and process time parameters; Acquire basic engineering data, and calculate the basic scheduling parameters for a single blasting operation based on the basic engineering data, specifically including the following steps: Based on the tunnel cross-sectional parameters and the blasting design parameters, the amount of rock blasted in a single blast is calculated. Based on the parameters of the amount of blasted rock and the vehicle's operational capacity, the required number of transport trucks and the time required to load and unload the waste rock are calculated. Based on the time for loading waste rock and the process time parameters, the total fixed process time is calculated. The basic scheduling parameters are input into the construction scheduling model to generate an initial vehicle scheduling instruction set. The construction scheduling model is configured to apply preset scheduling rules and use a comprehensive optimization algorithm to solve for the total scheduling period, scheduling cost, and construction safety. Specifically, the steps include: Based on the basic scheduling parameters, the preset scheduling rules are applied to generate vehicle scheduling logic; the preset scheduling rules include: vehicle process matching rules and vehicle path priority rules. Taking the overall scheduling period, scheduling cost, and construction safety as comprehensive optimization objectives, the vehicle scheduling logic is solved through an optimization algorithm, and the globally optimized initial vehicle scheduling instruction set is output.

2. The optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method according to claim 1, characterized in that, Before generating vehicle scheduling logic based on the basic scheduling parameters and the preset scheduling rules, the preset scheduling rules include vehicle process matching rules and vehicle path priority rules, and the following steps are also included: Construct a spatial topology network for the tunnels and number the intersections, construction sections, vehicle types, and process statuses within the tunnels.

3. The optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method according to claim 2, characterized in that, The vehicle process matching rule is implemented through a destination function. The input of the destination function is the process type number and the construction section number, and the output of the destination function is the vehicle type number that should be dispatched for the corresponding process type.

4. The optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method according to claim 1, characterized in that, The vehicle route priority rule is implemented through a dynamic priority function, which assigns priority based on the vehicle's load status, with loaded vehicles having a higher priority than unloaded vehicles.

5. The optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method according to claim 1, characterized in that, Taking the overall scheduling period, scheduling cost, and construction safety as comprehensive optimization objectives, the vehicle scheduling logic is solved using an optimization algorithm to output the globally optimized initial vehicle scheduling instruction set. Specifically, the steps include: A multi-objective fitness function is constructed to comprehensively evaluate any vehicle scheduling scheme; the multi-objective fitness function is a weighted sum based on the total scheduling time function, scheduling cost function, and safety score function; Using the multi-objective fitness function as the evaluation criterion, the optimization algorithm is run to iteratively search and compare various vehicle scheduling schemes generated based on the vehicle scheduling logic, and the vehicle scheduling scheme with the optimal fitness value is output as the initial vehicle scheduling instruction set.

6. The optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method according to claim 5, characterized in that, Using the multi-objective fitness function as the evaluation criterion, the optimization algorithm is run to iteratively search and compare various vehicle scheduling schemes generated based on the vehicle scheduling logic, and the vehicle scheduling scheme that optimizes the multi-objective fitness function value is output as the initial vehicle scheduling instruction set. Specifically, the process includes the following steps: S1: Encode different vehicle scheduling schemes to generate an initial population containing multiple vehicle scheduling schemes; S2: Using the multi-objective fitness function as the evaluation criterion, repeatedly perform selection, crossover, and mutation operations on the initial population to generate the next generation population; S3: Each time, use the generated next generation population as a new initial population and repeat step S2 until the iteration termination condition is met. S4: Decode the individual with the best fitness value in the population obtained when the termination condition is met, and output it as the initial vehicle scheduling instruction set.

7. The optimized scheduling method for construction vehicles applicable to underground construction using the drill-and-blast method according to claim 1, characterized in that, When a dynamic scheduling event is detected, a rescheduling mechanism is triggered. Based on the current construction status and vehicle operation status, the input parameters of the construction scheduling model are updated, and the construction scheduling model is re-solved to generate an updated vehicle scheduling instruction set to control the construction vehicles to continue performing operations. Specifically, the following steps are included: By comparing the monitored construction status with the vehicle operation status and the currently executed vehicle dispatch instruction set, it is determined whether a dynamic dispatch event has occurred; the dynamic dispatch event includes the vehicle staying at a critical node for an extended period of time, the actual time consumed by the process exceeding the planned threshold, or the vehicle reporting a fault. When the dynamic scheduling event is determined to have occurred, the rescheduling mechanism is triggered; Based on the current construction status and vehicle operation status, update the input parameters of the construction scheduling model. The input parameters include at least: the location of affected vehicles, the status of blocked processes, and available equipment resources. The input parameters are input into the construction scheduling model to control the optimization algorithm to solve the problem again and generate an updated vehicle scheduling instruction set. The updated vehicle dispatch instruction set is sent to the corresponding construction vehicles to control the vehicles to continue performing operations.

Images