Strip mine unmanned mine card dynamic scheduling method combining rules and models

Abstract

The invention relates to the technical field of unmanned driving, in particular to a strip mine unmanned mine card dynamic scheduling method combining rules and models, and the method comprises the steps: obtaining a reachable scheduling target of an unmanned mine card; if only one reachable scheduling target exists, determining the reachable scheduling target as an optimal scheduling target; if a plurality of reachable scheduling targets exist, simulating the production process of the unmanned mine card based on the plurality of reachable scheduling targets by adopting a model-based scheduling method to obtain a calculation result of the model-based scheduling method; judging whether the calculation result of the model-based scheduling method has calculation timeout or calculation failure or not, and determining an optimal scheduling target according to the judgment result of the calculation timeout or the calculation failure; scheduling the unmanned mine cards according to the optimal scheduling target; according to the method, the calculation efficiency and stability of dynamic scheduling of the unmanned mine cards can be improved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned driving technology, and in particular to a method for dynamically dispatching unmanned mining trucks in an open-pit mine in combination with rules and models. Background Art

[0002] Dynamic scheduling of unmanned mining trucks in open-pit mines is an important research direction in the field of mine automation, aiming to improve the overall efficiency and safety of mining operations. At present, many mining companies mainly use scheduling solutions based on fixed rules in the process of scheduling unmanned mining trucks. Although these rule-driven solutions can work effectively in some simple scenarios, their limitations in complex scenarios have become increasingly apparent.

[0003] Traditional rule-based scheduling methods rely on pre-set scheduling rules. For example, when a mining truck needs to select a loading location, the system usually determines the optimal scheduling target based on the "shortest distance" principle. This method is highly efficient in single-target situations, but in complex situations where multiple shovels have similar loading situations, it is easy to lead to suboptimal or even wrong scheduling decisions, thereby reducing the effectiveness of the overall scheduling.

[0004] In order to overcome the above shortcomings, some researchers have proposed model-based intelligent scheduling algorithms. This type of algorithm can calculate a more reasonable scheduling plan in a complex environment by modeling and analyzing the mining environment and the operating status of mining trucks. However, there are many problems in practical applications of scheduling algorithms that rely solely on models. First, the construction and operation of the model usually consume a lot of computing resources and time costs, and may not be able to generate scheduling results in time in an emergency. Secondly, when dealing with certain special situations, model-based algorithms may fail to calculate or the results may be unstable. Summary of the invention

[0005] In view of the above problems, the present invention provides a dynamic scheduling method for unmanned mining trucks in open-pit mines that combines rules and models, which solves the technical problems of low calculation efficiency and unstable results of model-based scheduling methods in the prior art.

[0006] The present invention provides a method for dynamic scheduling of unmanned mining trucks in open-pit mines combining rules and models, comprising the following steps:

[0007] Step S1, based on the idle positions in multiple mine production stages, obtaining the reachable scheduling target of the unmanned mining truck;

[0008] Step S2, determining the number of the reachable scheduling targets, if there is only one reachable scheduling target, determining the reachable scheduling target as the optimal scheduling target; if there are multiple reachable scheduling targets, adopting a model-based scheduling method, simulating the production process of the unmanned mining truck based on the multiple reachable scheduling targets, and obtaining the calculation result of the model-based scheduling method;

[0009] Step S3, judging whether the calculation result of the model-based scheduling method has a calculation timeout or a calculation failure, and determining the optimal scheduling target according to the judgment result of the calculation timeout or the calculation failure;

[0010] Step S4: dispatching the unmanned mining truck according to the optimal dispatching target.

[0011] Preferably, step S1 specifically includes: determining in sequence whether the number of free parking spaces, free pre-parking spaces, free parking areas and dispatched vehicles is greater than zero, and determining a reachable dispatch target of the unmanned mining truck from the free parking spaces, free pre-parking spaces, free parking areas and dispatched vehicles.

[0012] Preferably, the step of determining the achievable dispatch target of the unmanned mining truck from the free parking spaces, the free pre-parking spaces, the free parking areas and the dispatched vehicles specifically comprises:

[0013] Determine the current number of free parking spaces, and if the number of free parking spaces is greater than 0, determine all free parking spaces as the reachable scheduling targets;

[0014] If the number of free parking spaces is 0, the current number of free pre-parking spaces is determined; if the number of free pre-parking spaces is greater than 0, all free pre-parking spaces are determined as the reachable scheduling targets;

[0015] If the number of free pre-parking spaces is 0, the number of current free parking areas is determined; if the number of free parking areas is greater than 0, all free parking areas are determined as the reachable scheduling targets;

[0016] If the number of the idle parking areas is 0, the number of currently dispatched vehicles is determined; if the number of dispatched vehicles is greater than 0, all dispatched vehicle locations are determined as the reachable scheduling targets.

[0017] Preferably, if there are multiple achievable scheduling targets in step S2, a model-based scheduling method is adopted to simulate the production process of the unmanned mining truck based on the multiple achievable scheduling targets, and the calculation results of the model-based scheduling method specifically include:

[0018] (1) determining that the number of the plurality of achievable scheduling targets is N, sampling the time-varying parameter estimation values ​​M times, and obtaining M groups of time-varying parameter estimation values;

[0019] (2) Based on the discrete simulation model, N*M production process simulations are performed on N achievable scheduling targets based on the M groups of time-varying parameter estimation values, and the number of times each achievable scheduling target is evaluated as the optimal solution in the M sampling results is counted;

[0020] (3) The achievable scheduling target that is rated as the optimal solution the most times is determined as the scheduling target output by the model-based scheduling method.

[0021] Preferably, the calculation result of the model-based scheduling method includes: calculation timeout, calculation failure or calculation success; when the calculation is successful, the model-based scheduling method outputs a scheduling target.

[0022] Preferably, the estimated values ​​of the time-varying parameters include an estimated value of the loading time and an estimated value of the stop time of the mine car;

[0023] Analyze the loading time and the entry and stop time of each mine car through a box plot to obtain the estimated value of the loading time and the estimated value of the entry and stop time;

[0024] The loading time is the difference between the loading end time and the loading start time; the parking time is the time it takes for the mine car to travel from the parking area of ​​the main road to the loading position.

[0025] Preferably, the simulation process of the discrete simulation model comprises the following steps:

[0026] (1) For each virtual mining truck in the virtual parking area queue, the passing process and the running process are performed in sequence, and the virtual pre-parking position is reached, and each virtual mining truck enters the virtual pre-parking position queue;

[0027] (2) For each virtual mining truck in the virtual pre-parking position queue, the passing process and the running process are performed in sequence, and the virtual mining truck goes to the virtual parking position, and each virtual mining truck enters the virtual parking position queue;

[0028] (3) For each virtual mining truck in the virtual parking space queue, the loading process and the running process are carried out in sequence, and the virtual mining truck goes to the virtual unloading area, and each virtual mining truck enters the virtual workbench queue.

[0029] Preferably, step S3 specifically includes: judging the calculation result of the model-based scheduling method, if the result is a calculation timeout or a calculation failure, determining the scheduling target closest to the unmanned mining truck as the optimal scheduling target; if the result is a calculation success, determining the scheduling target output by the model-based scheduling method as the optimal scheduling target.

[0030] Compared with the prior art, the present invention has at least the following beneficial effects:

[0031] (1) By combining the rule-based scheduling process with the model-based intelligent scheduling algorithm, the present invention can quickly output the scheduling results when a single optimal scheduling target is clear. In complex situations where multiple electric shovels have similar loading conditions, the intelligent scheduling algorithm is used to calculate the optimal scheduling solution, thereby significantly improving the scheduling accuracy and overall operating efficiency.

[0032] (2) Based on the rule-based scheduling scheme, the present invention integrates an intelligent scheduling algorithm to effectively deal with various emergencies in complex environments. Through the mutual complementation of rule and model scheduling methods, the system can adopt the rule scheduling based on the closest distance when the model calculation times out or fails, ensuring the stable output of the scheduling scheme, thereby improving the stability and robustness of the system.

[0033] (3) In simple operation situations, the present invention gives priority to the rule-based fast scheduling method and only calls the model-based intelligent scheduling algorithm in complex situations, thereby reducing unnecessary computing resource consumption and time costs, avoiding resource waste and scheduling conflicts, optimizing the configuration of resources such as mining trucks and excavators, and improving the overall efficiency of mining operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The drawings are only for the purpose of illustrating particular embodiments and are not to be construed as limiting the invention.

[0035] Figure 1 A flow chart of a method for dynamic scheduling of unmanned mining trucks in open-pit mines combining rules and models provided by the present invention.

[0036] Figure 2 A flowchart of a scheduling algorithm based on fixed rules provided by the present invention.

[0037] Figure 3 Schematic diagram of the scheduling system framework based on the “rule + model” fusion provided by the present invention.

[0038] Figure 4 A block diagram of the intelligent scheduling method based on simulation tools provided by the present invention.

[0039] Figure 5 A block diagram of a discrete event-driven simulation model provided by the present invention. DETAILED DESCRIPTION

[0040] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. In addition, the present invention can also be implemented in other ways different from those described herein, and therefore, the protection scope of the present invention is not limited by the specific embodiments disclosed below.

[0041] Traditional rule-based scheduling methods such as Figure 2 As shown, it depends on the pre-set scheduling rules. The method provided by the present invention combines the rule-based main process scheme with the model-based intelligent scheduling algorithm, aiming to achieve rapid judgment of a single optimal scheduling target, while solving the optimal scheduling selection problem when multiple electric shovels have similar loading situations. Through the "rule + model" fusion scheduling scheme, the present invention not only improves the scheduling efficiency and accuracy, but also enhances the stability and robustness of the system.

[0042] The method provided by the present invention directly outputs the excavator as the optimal scheduling target when the system determines that only one excavator has an idle loading position; if all excavators are in loading operation, the system determines the pre-parking position, and the judgment logic of the pre-parking position is consistent with the parking position; when multiple excavators have idle loading positions, the system first selects the excavator closest to the distance according to a fixed rule. If the effectiveness of this method is limited, the system calls the model-based intelligent scheduling algorithm. The algorithm provides two result return mechanisms: if the calculation times out or fails, it is selected according to the closest distance; if the calculation is successful, the optimal scheduling target is directly output.

[0043] In order to illustrate the effectiveness of the method proposed by the present invention, the above technical solution of the present invention is described in detail below through a specific embodiment. Figure 1 As shown, a method for dynamic scheduling of unmanned mining trucks in open-pit mines combining rules and models is disclosed, and the specific implementation steps are as follows:

[0044] Step S1: based on the idle positions in multiple mine production stages, obtain the achievable scheduling target of the unmanned mining truck.

[0045] like Figure 3 The figure shows the processing process of the rule-based dispatching method of the present invention. During the processing, the number of free parking spaces, free pre-parking spaces, free parking areas and dispatched vehicles is judged in turn to determine the achievable dispatching target of the unmanned mining truck. The specific steps are:

[0046] Determine the current number of free parking spaces, and if the number of free parking spaces is greater than 0, determine all free parking spaces as the reachable scheduling targets;

[0047] If the number of free parking spaces is 0, the current number of free pre-parking spaces is determined; if the number of free pre-parking spaces is greater than 0, all free pre-parking spaces are determined as the reachable scheduling targets;

[0048] If the number of free pre-parking spaces is 0, the number of current free parking areas is determined; if the number of free parking areas is greater than 0, all free parking areas are determined as the reachable scheduling targets;

[0049] If the number of the idle parking areas is 0, the number of currently dispatched vehicles is determined; if the number of dispatched vehicles is greater than 0, all dispatched vehicle locations are determined as the reachable scheduling targets.

[0050] Through the above scheduling rules, the present invention judges each stage of mine production through certain rules and finally determines one or more achievable scheduling targets.

[0051] Step S2: determine the number of reachable scheduling targets. If there is only one reachable scheduling target, determine the reachable scheduling target as the optimal scheduling target.

[0052] If there are multiple achievable scheduling targets, a model-based scheduling method is used to simulate the production process of the unmanned mining truck based on the multiple achievable scheduling targets to obtain a calculation result of the model-based scheduling method;

[0053] When there are multiple reachable scheduling targets, the existing scheme will select the closest excavator according to the established rules. However, in this case, when the loading conditions of multiple electric shovels are similar, this method is prone to suboptimal or even wrong scheduling selection, thereby reducing the overall scheduling efficiency and work performance. In addition, the scheduling method that relies solely on fixed rules cannot meet the needs of engineering applications when dealing with various emergencies, resulting in insufficient adaptability and flexibility of the scheduling system in complex scenarios, thereby affecting the operating efficiency and safety of unmanned mining trucks.

[0054] The present invention provides a scheduling solution that integrates "rules + models" and calls a model-based intelligent scheduling algorithm to overcome the limitations of the rule-based method. Figure 4 The figure shows the process of simulating the production process of unmanned mining trucks based on the model-based scheduling method.

[0055] When there are multiple achievable scheduling targets, first, the number of the multiple achievable scheduling targets is determined to be N, and the estimated value of the time-varying parameter is sampled M times. Based on the discrete simulation model, N*M production process simulations are performed, and the number of times each achievable scheduling target is rated as the optimal solution in the M sampling results is counted. The achievable scheduling target that is rated as the optimal solution the most times is determined as the scheduling target output by the model-based scheduling method.

[0056] The time-varying parameters may include the loading time and the parking time of the mine car, wherein the parking time refers to the time it takes for the mine car to travel from the parking area of ​​the main road to the loading position.

[0057] The estimation of time-varying parameters is based on historical data. The present invention collects 43,318 loading data, and performs membership fitting and analysis of related parameters based on these data. First, the loading time of the mine car shovel is analyzed. The loading time of the mine car is the difference between the loading end time and the loading start time in each data. The distribution of loading time of all shovels is shown by a box plot, and all loading times are fitted using a logarithmic exponential model. The results show that the parameters μ = ln (3.98), σ = 0.34. In practical applications, if the target shovel ID exists in the historical database, it is sampled from its data; otherwise, it is sampled from the overall fitted data.

[0058] Secondly, the entry and stop time of the electric shovel is analyzed. The entry and stop time includes the waiting time. Since the waiting time is difficult to analyze, this project analyzes the loading time of the mining truck that enters the loading area in advance to obtain the waiting time. The entry and stop time of each electric shovel is analyzed through a box plot.

[0059] The calculation of the model-based scheduling method of the present invention is time-limited, so the calculation results of the model-based scheduling method will include calculation timeout, calculation failure or calculation success, etc. When the calculation is successful, the system will output the scheduling target.

[0060] The discrete simulation model in the model-based scheduling method is a mine production model, also known as a rapid simulation tool or an open-pit mine digital twin system, whose main goal is to quickly and realistically simulate the production process of the open-pit mine.

[0061] The simulation process of the discrete simulation model is as follows: Figure 5 As shown in the figure, the simulation process includes the circulation process of mining trucks in different production stages. The specific steps are as follows:

[0062] For each virtual mining truck in the virtual parking area queue, the passing process and the moving process are carried out in turn, and the virtual pre-parking position is reached, and each virtual mining truck enters the virtual pre-parking position queue;

[0063] For each virtual mining truck in the virtual pre-parking position queue, the passing process and the running process are performed in sequence, and the virtual mining truck goes to the virtual parking position, and each virtual mining truck enters the virtual parking position queue;

[0064] For each virtual mining truck in the virtual parking space queue, the loading process and the running process are carried out in sequence, and the truck goes to the virtual unloading area, and each virtual mining truck enters the virtual workbench queue.

[0065] Through the above simulation process, the actual operating situation of the loading area is simulated, which can improve the accuracy of scheduling and the overall operating efficiency.

[0066] Step S3: determine whether a calculation result of the model-based scheduling method has a calculation timeout or a calculation failure, and determine an optimal scheduling target according to the judgment result of the calculation timeout or the calculation failure.

[0067] The specific judgment method is: judging the calculation result of the model-based scheduling method, if the result is a calculation timeout or a calculation failure, then determining the scheduling target closest to the unmanned mining truck as the optimal scheduling target; if the result is a calculation success, then determining the scheduling target output by the model-based scheduling method as the optimal scheduling target.

[0068] Through the above steps, the present invention integrates a model-based intelligent scheduling algorithm to effectively deal with various emergencies in complex environments. Through the mutual complementation of rule and model scheduling, the system can use the closest distance-based rule scheduling when the model calculation times out or fails to ensure the stable output of the scheduling solution, thereby improving the stability and robustness of the system.

[0069] Step S4: dispatching the unmanned mining truck according to the optimal dispatching target.

[0070] Through the above-mentioned scheduling system framework based on the fusion of "rules + models" and the intelligent scheduling algorithm based on discrete simulation models, the present invention effectively overcomes the limitations of the existing technology in complex environments, significantly improves the efficiency and accuracy of unmanned mining truck scheduling in open-pit mines, enhances the stability and robustness of the system, optimizes resource allocation and computing costs, and provides strong technical support for mine automation operations.

[0071] Although the specific embodiments of the present invention have been described in a specific order, each action or step should be understood to require that such action or step be performed in the specific order shown or in a sequential order, or require that all illustrated actions or steps should be performed to obtain the desired result. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Certain features described in the context of a separate embodiment can also be implemented in a single implementation in combination. On the contrary, the various features described in the context of a single implementation can also be implemented in multiple implementations individually or in any suitable sub-combination.

[0072] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for dynamic dispatching of unmanned mining trucks in open-pit mines combining rules and models, characterized in that: The following steps are involved: Step S1, based on the idle positions in multiple mine production stages, obtaining the reachable scheduling target of the unmanned mining truck; Step S2, determining the number of the reachable scheduling targets, if there is only one reachable scheduling target, determining the reachable scheduling target as the optimal scheduling target; if there are multiple reachable scheduling targets, adopting a model-based scheduling method, simulating the production process of the unmanned mining truck based on the multiple reachable scheduling targets, and obtaining the calculation result of the model-based scheduling method; Step S3, judging whether the calculation result of the model-based scheduling method has a calculation timeout or a calculation failure, and determining the optimal scheduling target according to the judgment result of the calculation timeout or the calculation failure; Step S4: dispatching the unmanned mining truck according to the optimal dispatching target.

2. The method for dynamic dispatching of unmanned mining trucks in open-pit mines combining rules and models according to claim 1 is characterized in that: Step S1 specifically includes: It is determined in turn whether the number of free parking spaces, free pre-parking spaces, free parking areas and dispatched vehicles is greater than zero, and the achievable dispatching target of the unmanned mining truck is determined from the free parking spaces, free pre-parking spaces, free parking areas and dispatched vehicles.

3. The method for dynamic dispatching of unmanned mining trucks in open-pit mines combining rules and models according to claim 2 is characterized in that: The step of determining the achievable dispatching target of the unmanned mining truck from the idle parking spaces, the idle pre-parking spaces, the idle parking areas and the dispatched vehicles specifically includes: Determine the current number of free parking spaces, and if the number of free parking spaces is greater than 0, determine all free parking spaces as the reachable scheduling targets; If the number of free parking spaces is 0, the current number of free pre-parking spaces is determined; if the number of free pre-parking spaces is greater than 0, all free pre-parking spaces are determined as the reachable scheduling targets; If the number of free pre-parking spaces is 0, the number of current free parking areas is determined; if the number of free parking areas is greater than 0, all free parking areas are determined as the reachable scheduling targets; If the number of the idle parking areas is 0, the number of currently dispatched vehicles is determined; if the number of dispatched vehicles is greater than 0, all dispatched vehicle locations are determined as the reachable scheduling targets.

4. The method for dynamic dispatching of unmanned mining trucks in open-pit mines combining rules and models according to claim 3 is characterized in that: If there are multiple achievable scheduling targets in step S2, a model-based scheduling method is used to simulate the production process of the unmanned mining truck based on the multiple achievable scheduling targets, and the calculation results of the model-based scheduling method specifically include: (1) determining that the number of the plurality of achievable scheduling targets is N, sampling the time-varying parameter estimation values ​​M times, and obtaining M groups of time-varying parameter estimation values; (2) Based on the discrete simulation model, N*M production process simulations are performed on N achievable scheduling targets based on the M groups of time-varying parameter estimation values, and the number of times each achievable scheduling target is evaluated as the optimal solution in the M sampling results is counted; (3) The achievable scheduling target that is rated as the optimal solution the most times is determined as the scheduling target output by the model-based scheduling method.

5. The method for dynamic dispatching of unmanned mining trucks in open-pit mines combining rules and models according to claim 4 is characterized by: The calculation results of the model-based scheduling method include: calculation timeout, calculation failure or calculation success; when the calculation is successful, the model-based scheduling method outputs a scheduling target.

6. The method for dynamic dispatching of unmanned mining trucks in open-pit mines combining rules and models according to claim 5 is characterized by: The estimated values ​​of the time-varying parameters include an estimated value of the loading time and an estimated value of the stop time of the mine car; Analyze the loading time and the entry and stop time of each mine car through a box plot to obtain the estimated value of the loading time and the estimated value of the entry and stop time; The loading time is the difference between the loading end time and the loading start time; the parking time is the time it takes for the mine car to travel from the parking area of ​​the main road to the loading position.

7. The method for dynamic dispatching of unmanned mining trucks in open-pit mines combining rules and models according to claim 6 is characterized in that: The simulation process of the discrete simulation model comprises the following steps: (1) For each virtual mining truck in the virtual parking area queue, the passing process and the running process are performed in sequence, and the virtual pre-parking position is reached, and each virtual mining truck enters the virtual pre-parking position queue; (2) For each virtual mining truck in the virtual pre-parking position queue, the passing process and the running process are performed in sequence, and the virtual mining truck goes to the virtual parking position, and each virtual mining truck enters the virtual parking position queue; (3) For each virtual mining truck in the virtual parking space queue, the loading process and the running process are carried out in sequence, and the virtual mining truck goes to the virtual unloading area, and each virtual mining truck enters the virtual workbench queue.

8. The method for dynamic dispatching of unmanned mining trucks in open-pit mines combining rules and models according to claim 7 is characterized in that: Step S3 specifically includes: The calculation result of the model-based scheduling method is judged. If the result is a calculation timeout or a calculation failure, the scheduling target that is closest to the unmanned mining truck is determined as the optimal scheduling target; if the result is a calculation success, the scheduling target output by the model-based scheduling method is determined as the optimal scheduling target.

Images