Intelligent dispatching method and system for underground mine ore removal shoveling and transporting equipment

Abstract

The invention relates to the technical field of equipment scheduling, and discloses an intelligent scheduling method and system for underground mine ore removal shoveling equipment, and the method comprises the steps: processing multi-source data in a mine equipment scheduling process, and obtaining the initial feature representation of each scheduling element; performing data processing on the initial feature representation to obtain multi-dimensional constraint data of mine scheduling, and constructing an objective function of an optimization algorithm; coding decision variables corresponding to the scheduling scheme into particle position vectors based on a target function, and calculating the comprehensive performance of each particle through a fitness function; the population is generated by adopting different initialization strategies, so that the quality of an initial solution is improved; in particle swarm optimization iteration, a depth deterministic strategy gradient algorithm is introduced to realize adaptive regulation and control of core parameters, and taboo search is adopted to avoid local optimum; a specific execution plan is generated according to the optimal scheduling scheme obtained through training, and intelligent scheduling is achieved; the problem that an existing equipment scheduling method is low in solving efficiency is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of device scheduling, in particular to a kind of underground mine mining shovel truck device intelligent scheduling method and system. BACKGROUND

[0002] In recent years, with the development of high-tech mining informatization construction, the intelligent level of mine construction is continuously improved, and higher requirements are put forward for the intelligent scheduling of underground ore flow. The current scheduling method mainly relies on artificial experience, and the operation mode is inefficient, which leads to the decline of production efficiency and the rise of production cost. Existing research shows that meta-heuristic algorithms represented by genetic algorithm and particle swarm optimization algorithm can effectively optimize the process operation sequence and device resource allocation in typical manufacturing system scheduling scenarios such as job shop scheduling, which provides a feasible technical approach for the intelligent scheduling method of underground ore flow. However, traditional meta-heuristic algorithms are mostly used for continuous space optimization, and additional discretization processing is required in discrete scheduling problems, which may lead to loss of solution accuracy or slow convergence speed. Secondly, the parameters such as inertia weight and learning factor of traditional algorithms are fixed values, which cannot be dynamically adjusted according to the complexity of the scheduling problem and the iteration process. In simple problems, it may waste computing power due to excessive iteration, and in complex problems, it may fall into local optimum and be difficult to find the global optimal scheduling scheme.

[0003] Therefore, it is urgent to solve the technical problem that the key parameters in the existing meta-heuristic algorithm cannot be dynamically adjusted, resulting in that the solution efficiency and solution speed cannot achieve the expected effect. SUMMARY

[0004] The present application provides a kind of underground mine mining shovel truck device intelligent scheduling method and system to solve the problem of low solution efficiency of existing device scheduling method.

[0005] In order to achieve the above purpose, the technical scheme is as follows: In the first aspect, the present application provides a kind of underground mine mining shovel truck device intelligent scheduling method, comprising: Obtain multi-source data in the process of mine device scheduling, and discretize and encode the multi-source data to obtain the initial feature representation of each scheduling element; Data processing is carried out on the initial feature representation to obtain multi-dimensional constraint data of mine device scheduling, and the objective function of optimization algorithm is constructed based on the multi-dimensional constraint data; Encode the decision variable corresponding to the scheduling scheme into a particle position vector based on the objective function, initialize the particle position vector to obtain an initial population, and update the initial population iteratively; In the initial population iterative update, a path reconnection algorithm is used to select candidate solutions to construct a tabu search candidate set, and a tabu search algorithm is used to optimize the quality of the population in the iterative update based on the tabu search candidate set to obtain an optimal solution. The particle position vector of the optimal solution is constructed by a deep deterministic gradient algorithm, and the comprehensive performance numerical value of each particle position vector is calculated by using the fitness function. The particle with the highest comprehensive performance numerical value is taken as the optimal scheduling scheme to generate a specific execution plan, realizing intelligent scheduling.

[0006] Optionally, multi-source data in the mine equipment scheduling process is obtained, including: The production equipment state information, material storage information and environmental safety parameters are collected, and the production equipment state information, material storage information and environmental safety parameters are integrated by time stamp alignment to obtain multi-source data; The multi-source data is discretely coded to obtain the initial feature representation of each scheduling element, including: The mine ore flow transportation process is abstracted as a workpiece processing process of flow shop scheduling, and two-stage machine sets are defined, wherein the two-stage machine sets include: a first-stage mining and loading stage machine set MC and a second-stage transportation stage machine set MD; The multi-source data is abstracted as a workpiece, and the processing order of all processes on each machine in the two stages is traversed to obtain a corresponding solution scheme, each solution scheme individual including m sub-sequences corresponding to the m machines in the stages; The solution scheme is discretely coded to obtain the initial feature representation, and the solution space corresponding to the discretely coded mode satisfies the following relationship: ; In the formula, indicates the number of stages, is the number of workpieces, is the number of machines in the i-th stage, indicates the factorial.

[0007] Optionally, a target function of an optimization algorithm is constructed based on the multi-dimensional constraint data, including: Based on the multi-dimensional constraint data, the optimization target is to minimize the transportation time, and the target function F of the optimization algorithm is constructed, which satisfies the following relationship: ; In the formula, indicates the end time of the i-th workpiece on the transportation stage machine, indicates the start time of the i-th workpiece on the mining and loading stage machine.

[0008] Optionally, the particle position vectors are population initialized to obtain an initial population, and the initial population is iteratively updated, including: Half of the particle position vectors are encoded by using a heuristic algorithm to obtain permutation coded individuals, and the other half of the particle position vectors are encoded by using a random strategy to obtain random coded individuals, and the initial population is obtained by integrating the permutation coded individuals and the random coded individuals; The particle updating mechanism is established to update each particle position vector to obtain an updated each particle position vector; The updating mechanism includes a velocity updating mechanism and a position updating mechanism, and the velocity updating mechanism satisfies the following relationship: ; The position updating mechanism satisfies the following relationship: ; In the formula, denotes the velocity vector of the particle i at the tth iteration, denotes the position of the particle i at the tth iteration, denotes the optimal position of the particle i at the tth iteration, denotes the global optimum of the particle group at the tth iteration, denotes the inertia weight, is an individual learning factor, is a social learning factor, and is a random number in [0, 1].

[0009] Optionally, in the initial population iterative updating, a path reconnection algorithm is used to select candidate solutions to construct a tabu search candidate set, including: Taking the current solution of the individual, the historical optimal solution of the individual, and the historical optimal solution of the population as core nodes, search paths from the current solution to the historical optimal solution of the individual and from the current solution to the historical optimal solution of the population are constructed; For each candidate solution on the search path, the sequence similarity S (P,Q) of the candidate solution with the current solution of the individual, the historical optimal solution of the individual, and the historical optimal solution of the population is calculated, respectively; A similarity threshold range is preset, and candidate solutions with similarity in the threshold range are screened out, and the screened candidate solutions are added to the tabu search candidate set; In the process of adding to the tabu search candidate set, Kendall-Tau distance is introduced, and given two sequences P and Q with a length of n, the Kendall-Tau distance is defined as follows: ; In the formula, denotes the i-th element of the sequence P, denotes the i-th element of the sequence Q, The logical AND operator, The operator representing logical OR, Indicates the Kendall-Tau distance; The similarity calculation formula is defined as follows: ; In the formula, n represents the number of elements in the sequence, and S(P,Q) is the similarity between the two sequences.

[0010] Optionally, the optimal solution can be obtained by using the tabu search algorithm to optimize the quality of the population during iterative updates based on the tabu search candidate set, including: The candidate solutions in the tabu search candidate set are initialized, and the candidate solutions are used as the starting search points for the tabu search algorithm. A neighborhood search is performed based on the starting search point to obtain an updated solution. The neighborhood search includes: Obtain the workpiece processing process corresponding to the candidate solution in the tabu search candidate set, identify the longest process in the workpiece processing as the key block, take turns using the steps in the process within the key block as the first or last step of the key block to obtain the neighborhood search process, determine the duration of the neighborhood search process, compare the duration of the neighborhood search process with the duration of the key block, and when the duration of the neighborhood search process is less than the duration of the key block, take the corresponding neighborhood search process as the updated new solution; We construct taboo and amnesty criteria to judge the updated solution. When the updated solution satisfies the amnesty criterion, it is taken as the optimal solution. When the updated solution does not satisfy the amnesty criterion, the optimal solution is selected from the solutions that are not taboo. Among them, the taboo and amnesty criteria include taboo criteria and amnesty criteria. The taboo criteria include: establishing a process taboo table through manual construction, comparing the neighborhood search process with the process taboo table, and when the neighborhood search process is in the process taboo table, the updated solution corresponding to the neighborhood search process is the taboo solution, and when the neighborhood search process is not in the process taboo table, the updated solution corresponding to the neighborhood search process is the untabbed solution. The amnesty criteria include: when the updated solution is better than the global optimal solution, it can be directly selected as the optimal solution without being affected by the taboo criteria; The global optimal solution is the candidate solution with the shortest process time in the tabu search candidate set. The condition for the updated solution to be better than the global optimal solution is that the process time of the updated solution is shorter than the process time of the global optimal solution.

[0011] Optionally, the particle position vector of the optimal solution is constructed using a depth deterministic gradient algorithm, including: The optimal solution is divided into a state set, which contains three types of core indicators. One type of core indicator is population diversity. The dispersion of particle distribution in the solution space is quantified by the standard deviation of the particle position vector. The calculation formula is as follows: ; In the formula, Indicates the iteration number t. This represents the value of the d-th dimension for the i-th individual, where n is the population size and D is the dimension. The second type of core indicator is the improvement rate of the optimal solution. The change in the global optimal fitness over k consecutive generations reflects the optimization efficiency. The calculation formula is: ; In the formula, This represents the optimal solution for the population at the t-th iteration, based on the optimal solution k generations ago; The three core metrics are convergence progress P: the ratio of the current iteration number to the total number of iterations, expressed by the formula: ; In the formula, This indicates the maximum number of iterations preset by the algorithm; Based on three core indicators, a set of states is constructed, and the parameters are normalized to the [-1, 1] interval using a mapping formula: ; In the formula, param represents the parameters included in the aforementioned three types of indicators, and the value of i ranges from 0 to 5. This represents the normalized mapping parameters; Based on the normalized mapping parameters, a 15-dimensional state input vector is finally constructed. Inertia weights, a key parameter for population iterative updates Cognitive acceleration coefficient Social acceleration coefficient To regulate the inertial weights, a neural network and reward function are constructed. Cognitive acceleration coefficient Social acceleration coefficient renew; Regarding the design of the reward function: used to calculate the reward value after performing a certain action, the reward function is designed as follows: ; In the formula, Indicates the first The optimal solution in the next iteration. This represents the reward value in the t-th iteration; For neural network design: Neural networks include actor networks. Target Actor Network , action value network , and target action value network ; To train the action value network, a loss function is minimized: ; Subsequently, the actor network is trained by the policy gradient using the action value network, and the corresponding policy gradient formula is as follows: ; wherein, is a discount factor, is a loss function of the action value network, is a policy gradient of the actor network, is a reward value in the tth iteration, is a target action value network in the t+1th iteration, is a gradient operator, is a parameter vector of the actor network, is an objective function of the actor network, is an action vector, is an action vector in the tth iteration, is a state vector, is a state vector in the tth iteration, is an output value of the action value network, is a gradient of the action value network, is a gradient of the actor network, is a parameter adjustment action output by the actor network based on the state vector in the tth iteration; an inertia weight , a cognitive acceleration coefficient , a social acceleration coefficient , and a 15-dimensional state input vector are input into the trained actor network to obtain an updated parameter inertia weight , a cognitive acceleration coefficient , a social acceleration coefficient ; an updated parameter inertia weight , a cognitive acceleration coefficient , a social acceleration coefficient are substituted into the particle updating mechanism to update and iterate the optimal solution to obtain a final solution; the final solution is input into the objective function to obtain a particle position vector of the optimal solution.

[0012] Optionally, a fitness function is used to calculate a comprehensive performance numerical value of each particle position vector, including: The target function F is calculated by inputting each particle position vector into the target function, and the comprehensive performance value of each particle position vector is calculated by substituting the target function F into the fitness function. The fitness function satisfies the following relationship: In the formula, F is the target function value, The comprehensive performance value of each particle position vector.

[0013] Optionally, the particle with the highest comprehensive performance value is used to generate a specific execution plan for the optimal scheduling scheme, including: Outputting the optimal code corresponding to the particle with the highest comprehensive performance value, and generating a flow Gantt chart based on the optimal code; Decoding the flow Gantt chart to obtain a specific execution plan.

[0014] In a second aspect, the embodiments of the present application provide an intelligent scheduling system for underground mine ore extraction and shovel truck equipment, including a processor and a memory. The memory is used to store computer programs. The processor is used to execute the programs stored in the memory, and implement the method steps of any one of the first aspect.

[0015] Advantages: The intelligent scheduling method for underground mine ore extraction and shovel truck equipment provided by the present application uses a reinforcement learning algorithm to intelligently update w, c1, and c2 in the discrete particle swarm algorithm, thereby enhancing the solution efficiency and rate. The state set of the chromosome is divided, and the average fitness and diversity of the population are used as the basis for reflecting the overall state of the population, which is beneficial to improving the population quality and more easily obtaining excellent individuals. The tabu search algorithm is introduced to avoid local convergence too quickly. The scheduling optimization algorithm proposed by the present application can significantly reduce the waiting time of vehicles when picking up goods, improve the throughput, indirectly shorten the maximum operation time of underground ore flow transportation, and meet the work efficiency requirements of underground ore flow transportation. The optimization algorithm proposed by the present application can solve the problems of slow response, low precision, and poor adaptability of traditional scheduling methods in complex scenarios, and provides a standardized and scalable technical solution for intelligent scheduling in various industries. It belongs to the frontiers of the cross and fusion of disciplines such as artificial intelligence, industrial engineering, transportation engineering, and logistics engineering. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The flowchart of the intelligent scheduling method for underground mine ore extraction and shovel truck equipment of the preferred embodiments of the present application is shown in the figure. Figure 2 The algorithm flowchart of the intelligent scheduling method for underground mine ore extraction and shovel truck equipment of the preferred embodiments of the present application is shown in the figure. Figure 3 ​The flow chart of the contraindicated search algorithm provided for the preferred embodiment of the present application is as follows: Figure 4 The schematic diagram of neural network training of the DDPG algorithm provided for the preferred embodiment of the present application is as follows. DETAILED DESCRIPTION

[0017] The technical solutions of the present application will be described below in a clear and complete manner. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0018] Unless otherwise defined, the technical terms or scientific terms used in the present application shall have the usual meanings understood by those skilled in the art to which the present application belongs. The terms "first", "second", and similar terms used in the present application do not represent any order, number, or importance, but are only used to distinguish different components. Similarly, the terms "one" or "a" and similar terms do not represent a quantity limitation, but represent the existence of at least one. The terms "connected" or "connected" and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms "up", "down", "left", "right", and the like are only used to represent relative positional relationships, and when the absolute positions of the described objects change, the relative positional relationships also change accordingly.

[0019] Please refer to Figures 1-2 The present application provides an intelligent scheduling method for underground mine ore extraction shovel truck equipment, comprising: S1, collecting multi-source data involved in the mine scheduling process, performing discretization coding processing, and obtaining initial feature representation of each scheduling element; In the preferred embodiment of the present application, data acquisition is intelligently collected through real-time sensors in three dimensions of production equipment state, material storage information, and environmental safety parameters. The three-level architecture of "sensor real-time acquisition + edge computing preprocessing + cloud aggregation" is adopted to realize data acquisition: the front end collects raw data through sensors, and transmits the raw data to the mine system through industrial Ethernet through the edge gateway, aligns the asynchronous data of multiple devices through time stamp, and finally forms a standardized raw data set.

[0020] S2, filtering processing is performed on the initial feature embedding obtained in S1 to remove data noise, and multi-dimensional constraints of mine scheduling are quantified, and an objective function of a particle swarm optimization algorithm is constructed; In the preferred embodiment of the present application, the machine loading efficiency and the stage operation sequence are considered, the process sequence constraint and the equipment capacity constraint are considered, and an optimization objective function F of minimizing the transportation time is constructed. The construction of the objective function is as follows:

[0021] In the formula, denotes the end time of the i th workpiece on the machine in the transportation stage, denotes the start time of the i th workpiece on the machine in the sampling and loading stage.

[0022] S3, the multi-objective function constructed based on S2, model training is carried out through a particle swarm optimization algorithm, and global optimization of the scheduling scheme is realized; In the preferred embodiment of the application, the specific form of the encoding mode is the processing sequence of all processes on each machine in each stage, and each solution scheme individual includes m sub-sequences, which correspond to m machines respectively, and the sequence in the sub-sequence is the sequence of process on the machine. Because each workpiece needs to flow through each stage, the sequence of all machines in each stage should contain all processes, and the solution space corresponding to the encoding mode is:

[0023] In the formula, denotes the number of stages, is the number of workpieces, is the number of machines in the i th stage, denotes factorial.

[0024] In the preferred embodiment of the application, after the individuals in the population are encoded by the specific encoding mode, the algorithm iteration and model training include: Population initialization, that is, four heuristic algorithms (SPTB, NEHLPT (λ), bLPTB, NEHbSPT (λ)) are used to generate half of the different characteristic high-quality permutation encoding individuals. In order to strengthen the diversity of the population, a random strategy is adopted to initialize the remaining individuals, which is beneficial to increase the difference between individuals in the population and ensure the diversity of the population. The generated permutation code is combined with the first available machine criterion to generate the encoding of the whole stage.

[0025] The fitness function of the individual is calculated to determine the current solution of the individual, the historical optimal solution of the individual, and the historical optimal solution of the population, and the fitness is calculated in the following form:

[0026] In the formula, is the value of the objective function, is the comprehensive performance value of each particle position vector.

[0027] The solution is updated through the particle swarm algorithm to form multiple particle position matrices and multiple particle velocity matrices, which are used to update the particle position and set the fitness function, wherein the particle position transformation and velocity change are as follows:

[0028] where, denotes the velocity vector of particle i at the tth iteration, denotes the position of particle i at the tth iteration, denotes the optimal position of particle i at the tth iteration, denotes the global optimum of the swarm at the tth iteration, denotes the inertia weight, is the individual learning factor, is the social learning factor, and is a random number in [0,1] for increasing randomness.

[0029] By the path reconnection algorithm, the search paths from the current solution to the individual historical optimal solution and from the current solution to the swarm historical optimal solution are constructed with the individual current solution, the individual historical optimal solution and the swarm historical optimal solution as the core nodes. Through specific selection, some solutions on the search paths are added to the tabu search candidate set, and the similarity between two solutions is obtained by calculating the Kendall-Tau distance (KTD).

[0030] In the preferred embodiment, for given two sequences P and Q, the length of which is n, the KTD distance is defined as follows:

[0031]

[0032] where, denotes the i th element of sequence P, denotes the i th element of sequence Q, denotes the logical and operator, denotes the logical or operator, denotes the Kendall-Tau distance, and the similarity calculation formula is defined as follows:

[0033] where n represents the number of elements of the sequence, and S (P,Q) is the similarity of the two sequences.

[0034] After formula conversion, the similarity is in the range of [0,1], which is convenient for controlling the similarity analysis of problems of different scales. When two sequences are completely consistent, the similarity is 1; when two sequences are completely opposite, the similarity is 0.

[0035] By specific selection of some solutions, the preferred solution is added to the candidate set, and quality optimization is performed by tabu search, such as Figure 3As shown, the tabu search algorithm is introduced to improve the local optimization ability and avoid premature convergence, as follows: The main steps of search include initialization, neighborhood search, tabu and exemption criteria determination and solution update. The initialization process starts from the candidate solution described above to provide the starting search point for the algorithm; the neighborhood search process generates new solutions in a specific way to expand the search space; the tabu and exemption criteria determination process helps to ensure the rationality of the search direction, and the new solution is better than the current optimal solution, which meets the exemption criteria, if it does not meet the exemption criteria, the optimal solution is selected from the non-tabu solution; the solution update process is to add the new solution to the current solution set and update the tabu list to continuously optimize the quality of the solution.

[0036] Specific neighborhood search includes: trying to move the process within all key blocks to before the block head or after the block tail, or inserting the key block head process or block tail process into the block; or also remove a key process from the current stage of the machine and insert it into another available machine in the same stage.

[0037] In the preferred embodiment, reinforcement learning is used to achieve adaptive control of key parameters, such as Figure 4 As shown, the main steps of reinforcement learning include: The deep deterministic gradient algorithm is introduced to achieve adaptive control of parameters: State set division: the state set contains three types of core indicators: (1) Population diversity : According to the standard deviation of the particle position vector, the distribution of the particle in the solution space is quantified. The formula is as follows:

[0038] Where, represents the tth iteration, represents the value of the ith individual in the dth dimension, n is the population size, and D is the dimension.

[0039] (2) Optimal solution improvement rate : The change of global optimal fitness in continuous k generations, reflecting the optimization efficiency.

[0040]

[0041] Where, represents the population optimal solution at the tth iteration, and the relative improvement ratio of the current optimal solution is calculated based on the optimal solution before k generations. represents performance improvement, represents improvement stagnation.

[0042] (3) Convergence progress P: The ratio of the current iteration number to the total number of iterations. It is used to indicate the iteration stage of the algorithm, reflecting whether the algorithm is in the exploration or development stage. The formula is shown below, where This indicates the maximum number of iterations preset by the algorithm.

[0043]

[0044] In the formula, This indicates the maximum number of iterations preset by the algorithm.

[0045] Based on the aforementioned three types of indicators, a set of states is constructed. Given that some parameters have extremely small values ​​and their dynamic changes are even more subtle, to enable the neural network to effectively perceive and learn these subtle changes, their values ​​are normalized to the interval [-1, 1] using the following mapping formula:

[0046] Here, param represents the parameters included in the aforementioned three types of indicators, and the value of i ranges from 0 to 5; based on this, a 15-dimensional state input vector is finally constructed.

[0047] Action set construction: using key parameters of the particle swarm optimization algorithm (inertia weight) Cognitive acceleration coefficient Social acceleration coefficient The algorithm dynamically optimizes parameter configurations based on the algorithm's iteration process, targeting parameters such as inertia weight and social acceleration coefficient. In the early iteration stage, it tends to increase inertia weight and decrease social acceleration coefficient to enhance the population's global search capability and avoid premature entrapment in local optima. In the middle iteration stage, it dynamically balances cognitive acceleration coefficient and social acceleration coefficient to achieve coordinated advancement of global exploration and local development, balancing solution diversity and optimization efficiency. In the later iteration stage, it decreases inertia weight and increases cognitive acceleration coefficient to guide the population to focus on the current optimal region for refined local optimization, improving solution accuracy. Based on these parameter control requirements, a four-dimensional action set is output through the Deep Deterministic Policy Gradient (DDPG) algorithm. This action set corresponds to the quantified adjustment signals of key parameters, providing precise control basis for the adaptive optimization of particle swarm optimization parameters.

[0048] Reward function design: Used to calculate the reward value after performing a certain action, its goal is to encourage the particle swarm optimization algorithm to obtain a better global optimum. The reward function is designed as follows:

[0049] in Indicates the first The optimal solution in the next iteration. This represents the reward value in the t-th iteration.

[0050] In the deep deterministic policy gradient algorithm, four neural networks are designed to obtain the optimal policy, namely the Actor Network. Target Actor Network Action Value Network (Critic Network) And the Target Critic Network. Among them, the Actor network With Target Actor Network Used for state-based action selection; Critic network With TargetCritic network This is used to evaluate the value of actions selected by the actor's network. To train the action value network, the following loss function needs to be minimized:

[0051] Subsequently, the Critic network is used to train the Actor network via policy gradient, and the corresponding policy gradient formula is as follows:

[0052] In the formula, As a discount factor, Let the loss function be the action value network. For the strategy gradient of the actor network, Let be the reward value in the t-th iteration. For the target action value network in the (t+1)th iteration, For gradient operators, For the parameter vector of the actor network, The objective function of the actor network, For action vectors, Let be the action vector for the t-th iteration. For state vectors, Let be the state vector for the t-th iteration. The output value of the action value network. For the gradient of the action value network, For the gradient of the actor network, For the parameter adjustment action of the actor network based on the state vector output at the t-th iteration.

[0053] Repeat the above algorithm iteration steps until the maximum number of iterations is reached. Output the historical best solution to generate a Gantt chart of the optimal scheduling scheme, and generate a specific execution plan to achieve intelligent scheduling.

[0054] The embodiment of the present application further provides an underground mine ore mining shovel-truck equipment intelligent scheduling system, comprising a processor, a memory; The memory is used for storing a computer program. The processor is used for executing the program stored on the memory, so as to realize the method steps of any one of the underground mine ore mining shovel-truck equipment intelligent scheduling method.

[0055] The underground mine ore mining shovel-truck equipment intelligent scheduling system can realize each embodiment of the underground mine ore mining shovel-truck equipment intelligent scheduling method, and achieve the same beneficial effects, which will not be described herein.

[0056] The preferred embodiments of the present application are described in detail above. It should be understood that those skilled in the art can make many modifications and variations without creative work according to the concept of the present application. Therefore, any technical solution obtained by logical analysis, reasoning or limited experiment on the basis of the prior art according to the concept of the present application shall be within the protection scope defined by the claims.

Claims

1. A method for intelligent scheduling of ore-extracting and transporting equipment in underground mines, characterized in that, include: Acquire multi-source data during the scheduling process of mining equipment, and perform discretization encoding on the multi-source data to obtain the initial feature representation of each scheduling element; The initial feature representation is processed to obtain multi-dimensional constraint data for mine equipment scheduling, and the objective function of the optimization algorithm is constructed based on the multi-dimensional constraint data. Based on the objective function, the decision variables corresponding to the scheduling scheme are encoded into particle position vectors. Population initialization is performed on the particle position vectors to obtain an initial population, and the initial population is iteratively updated. In the initial population iteration update, the path reconnection algorithm is used to select candidate solutions and construct a tabu search candidate set. The tabu search algorithm is then used to optimize the quality of the population in the iteration update based on the tabu search candidate set to obtain the optimal solution. The optimal particle position vector is constructed using a deep deterministic gradient algorithm, and the comprehensive performance value of each particle position vector is calculated using a fitness function. The particle with the highest overall performance value is used as the optimal scheduling scheme to generate a specific execution plan, thereby achieving intelligent scheduling.

2. The intelligent scheduling method for underground mine ore extraction and transportation equipment according to claim 1, characterized in that, Acquire multi-source data during the mine equipment scheduling process, including: Collect production equipment status information, material storage information, and environmental safety parameters, and integrate the production equipment status information, material storage information, and environmental safety parameters through timestamp alignment to obtain multi-source data; Discretizing and encoding the multi-source data yields initial feature representations for each scheduling element, including: The mining ore flow transportation process is abstracted as a workpiece processing process of a flow workshop. A two-stage machine set is defined for the processing process, which includes: the machine set MC for the first stage of mining and loading and the machine set MD for the second stage of transportation. The multi-source data is abstracted into workpieces. The two stages are traversed according to the processing order of all processes on each machine in the two stages to obtain the corresponding solution scheme. Each solution scheme includes m sub-sequences, which correspond to m machines in the stage respectively. The solution scheme is discretized to obtain the initial feature representation. The solution space corresponding to the discretization encoding method satisfies the following relationship: ； In the formula, Indicates the number of stages. For the number of workpieces, Let be the number of machines in the i-th stage. It represents factorial.

3. The intelligent scheduling method for underground mine ore extraction and transportation equipment according to claim 1, characterized in that, The objective function of the optimization algorithm is constructed based on the multi-dimensional constraint data, including: Based on multi-dimensional constraint data, with minimizing transportation time as the optimization objective, an objective function F of the optimization algorithm is constructed, which satisfies the following relationship: ； In the formula, This represents the end time of the i-th workpiece on the machine during the transportation phase. This indicates the start time of the i-th workpiece on the machine during the loading and unloading phase.

4. The intelligent scheduling method for underground mine ore extraction and transportation equipment according to claim 1, characterized in that, The particle position vectors are initialized to obtain an initial population, and the initial population is iteratively updated, including: A heuristic algorithm is used to encode half of the particle position vectors to obtain permuted coded individuals, and a random strategy is used to encode the other half of the particle position vectors to obtain randomly coded individuals. The permuted coded individuals and the randomly coded individuals are combined to obtain the initial population. A particle update mechanism is established to update the position vector of each particle, resulting in the updated position vector of each particle. The update mechanisms include: a speed update mechanism and a position update mechanism. The speed update mechanism satisfies the following relationship: ； The location update mechanism satisfies the following relationship: ； In the formula, This represents the velocity vector of particle i at the (t+1)th iteration. This represents the velocity vector of particle i at the t-th iteration. This represents the position of particle i in the (t+1)th iteration. This represents the position of particle i at the t-th iteration. Let represent the optimal position of particle i in the t-th iteration. This indicates the global optimum of the particle swarm at iteration t. Indicates inertia weight, For individual learning factors, As a social learning factor, and It is a random number in [0,1].

5. The intelligent scheduling method for underground mine ore extraction and transportation equipment according to claim 1, characterized in that, In the initial population iterative update, the path reconnection algorithm is used to select candidate solutions and construct a tabu search candidate set, including: Using the individual's current solution, the individual's historical best solution, and the population's historical best solution as core nodes, we construct search paths from the current solution to the individual's historical best solution and from the current solution to the population's historical best solution. For each candidate solution on the search path, calculate its sequence similarity S(P,Q) with the individual's current solution, the individual's historical best solution, and the population's historical best solution. A preset similarity threshold range is set, and candidate solutions with similarity within this threshold range are selected and added to the tabu search candidate set. The Kendall-Tau distance is introduced during the process of adding sequences to the tabu search candidate set. Given two sequences P and Q of length n, the Kendall-Tau distance is defined as follows: ； In the formula, This represents the i-th element of sequence P. This represents the i-th element of sequence Q. The logical AND operator, The operator representing logical OR, Indicates the Kendall-Tau distance; The similarity calculation formula is defined as follows: ； In the formula, n represents the number of elements in the sequence, and S(P,Q) is the similarity between the two sequences.

6. The intelligent scheduling method for underground mine ore extraction and transportation equipment according to claim 1, characterized in that, The optimal solution is obtained by using the tabu search algorithm to optimize the quality of the population during iterative updates based on the tabu search candidate set, including: The candidate solutions in the tabu search candidate set are initialized, and the candidate solutions are used as the starting search points for the tabu search algorithm. A neighborhood search is performed based on the starting search point to obtain an updated solution. The neighborhood search includes: Obtain the workpiece processing process corresponding to the candidate solution in the tabu search candidate set, identify the longest process in the workpiece processing as the key block, take turns using the steps in the process within the key block as the first or last step of the key block to obtain the neighborhood search process, determine the duration of the neighborhood search process, compare the duration of the neighborhood search process with the duration of the key block, and when the duration of the neighborhood search process is less than the duration of the key block, take the corresponding neighborhood search process as the updated new solution; We construct taboo and amnesty criteria to judge the updated solution. When the updated solution satisfies the amnesty criterion, it is taken as the optimal solution. When the updated solution does not satisfy the amnesty criterion, the optimal solution is selected from the solutions that are not taboo. Among them, the taboo and amnesty criteria include taboo criteria and amnesty criteria. The taboo criteria include: establishing a process taboo table through manual construction, comparing the neighborhood search process with the process taboo table, and when the neighborhood search process is in the process taboo table, the updated solution corresponding to the neighborhood search process is the taboo solution, and when the neighborhood search process is not in the process taboo table, the updated solution corresponding to the neighborhood search process is the untabbed solution. The amnesty criteria include: when the updated solution is better than the global optimal solution, it can be directly selected as the optimal solution without being affected by the taboo criteria; The global optimal solution is the candidate solution with the shortest process time in the tabu search candidate set. The condition for the updated solution to be better than the global optimal solution is that the process time of the updated solution is shorter than the process time of the global optimal solution.

7. The intelligent scheduling method for underground mine ore extraction and transportation equipment according to claim 1, characterized in that, The particle position vector for constructing the optimal solution is generated using a deep deterministic gradient algorithm, including: The optimal solution is divided into a state set, which contains three types of core indicators. One type of core indicator is population diversity. The dispersion of particle distribution in the solution space is quantified by the standard deviation of the particle position vector. The calculation formula is as follows: ； In the formula, Indicates the iteration number t. This represents the value of the d-th dimension for the i-th individual, where n is the population size and D is the dimension. The second type of core indicator is the improvement rate of the optimal solution. The change in the global optimal fitness over k consecutive generations reflects the optimization efficiency. The calculation formula is: ； In the formula, This represents the population-optimal solution at the t-th iteration. This represents the optimal solution for the population at the tk-th iteration, with the optimal solution from k generations ago as the benchmark. The three core metrics are convergence progress P: the ratio of the current iteration number to the total number of iterations, expressed by the formula: ； In the formula, This indicates the maximum number of iterations preset by the algorithm; Based on three core indicators, a set of states is constructed, and the parameters are normalized to the [-1, 1] interval using a mapping formula: ； In the formula, param represents the parameters included in the aforementioned three types of indicators, and the value of i ranges from 0 to 5. This represents the normalized mapping parameters; Based on the normalized mapping parameters, a 15-dimensional state input vector is finally constructed. Inertia weights, a key parameter for population iterative updates Cognitive acceleration coefficient Social acceleration coefficient To regulate the inertial weights, a neural network and reward function are constructed. Cognitive acceleration coefficient Social acceleration coefficient renew; Regarding the design of the reward function: used to calculate the reward value after performing a certain action, the reward function is designed as follows: ； In the formula, Indicates the first The optimal solution in the next iteration. Indicates the first The optimal solution in +1 iterations This represents the reward value in the t-th iteration; For neural network design: Neural networks include actor networks. Target Actor Network Action Value Network and target action value network ; To train the action value network, minimize the loss function: ； Subsequently, the actor network is trained using the action value network via policy gradient, and the corresponding policy gradient formula is as follows: ； In the formula, As a discount factor, Let the loss function be the action value network. For the strategy gradient of the actor network, Let be the reward value in the t-th iteration. For the target action value network in the (t+1)th iteration, For gradient operators, For the parameter vector of the actor network, The objective function of the actor network, For action vectors, Let t be the action vector for the t-th iteration. For state vectors, Let be the state vector for the t-th iteration. The output value of the action value network. For the gradient of the action value network, For the gradient of the actor network, The parameter adjustment action of the actor network based on the state vector output at the t-th iteration; Inertial weight Cognitive acceleration coefficient Social acceleration coefficient The updated parameters and inertial weights are obtained by inputting the 15-dimensional state input vector into the trained actor network. Cognitive acceleration coefficient Social acceleration coefficient ; Updated parameter inertia weights Cognitive acceleration coefficient Social acceleration coefficient Substitute the particle update mechanism to update and iterate the optimal solution to obtain the final solution; The final solution is input into the objective function to obtain the particle position vector of the optimal solution.

8. The intelligent scheduling method for underground mine ore extraction and transportation equipment according to claim 1, characterized in that, The overall performance value of each particle's position vector is calculated using the fitness function, including: Input each particle position vector into the objective function, calculate the objective function F, and then substitute the objective function F into the fitness function to calculate the comprehensive performance value of each particle position vector; The fitness function satisfies the following relationship: ； In the formula, The objective function value, This represents the overall performance value for each particle's position vector.

9. The intelligent scheduling method for underground mine ore extraction and transportation equipment according to claim 1, characterized in that, The particle with the highest overall performance value is used as the optimal scheduling scheme to generate a specific execution plan, including: Output the optimal code corresponding to the particle with the highest overall performance value, and generate a Gantt chart based on the optimal code; The Gantt chart of the process is decoded and transformed into a specific execution plan.

10. An intelligent scheduling system for ore-carrying equipment in underground mines, characterized in that, Including processor and memory; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the steps of the method described in any one of claims 1-9.

Images