Ore extraction control method and system

Abstract

The invention discloses an ore extraction control method and system, and aims to solve the problems of low efficiency, delay and insufficient accuracy in the ore extraction process. The method comprises the steps that an end side collects and analyzes ore data; performing deep analysis on the data uploaded by the end-to-end side to generate an optimization control instruction; the cloud carries out deep analysis and optimization on ore characteristics and a production chain; according to ore characteristics and real-time feedback, production equipment parameters are adjusted, the cloud end trains and optimizes a large model based on production data, and the optimized parameters are updated to the edge end and the end side; through a collaboration mechanism, distributing a global optimization task to a cloud end, distributing a quick response task to an edge end, distributing a low delay task to an end side, and dynamically adjusting task distribution; the invention further comprises a corresponding system. Through a multi-level cooperation mechanism, efficient data processing, accurate decision and dynamic control in the mining process are achieved, the intelligent level of ore mining is improved, and the resource utilization rate is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of ore mining, and in particular to an ore mining control method and system. Background Art

[0002] Ore mining is a dynamic and complex process, involving real-time changes in ore characteristics, equipment operating status, and environmental conditions. As mining continues to move toward automation and intelligentization, key processes like ore crushing and screening are placing higher demands on intelligent equipment control.

[0003] Mines are often located in remote areas, where network bandwidth and computing resources are significantly limited. This environment poses a significant challenge to the real-time and high-precision capabilities of traditional control methods. While relying solely on large cloud-based models for inference can provide high accuracy, it struggles to meet the real-time requirements of on-site production due to high network transmission latency. Deploying lightweight models solely on end-devices, while reducing latency, has limited ability to process complex ore characteristics and struggles to adapt to the complex and changing conditions of mining operations. Summary of the Invention

[0004] The present invention solves the problems of inefficiency, delay and lack of accuracy in the ore mining process, and proposes an ore mining control method and system. Through a multi-level collaborative mechanism, it realizes efficient data processing, accurate decision-making and dynamic control in the mining process, improves the intelligence level of ore mining and improves resource utilization.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: a method for controlling ore mining, comprising the following steps: S1: The end-side collects and analyzes ore data; the edge performs in-depth analysis of the data uploaded by the end-side to generate optimized control instructions; the cloud performs in-depth analysis and optimization of ore characteristics and production chains; S2 adjusts production equipment parameters based on ore characteristics and real-time feedback. The cloud trains and optimizes a large model based on production data, and updates the optimized parameters to the edge and end-points. S3, through a collaborative mechanism, allocates global optimization tasks to the cloud, fast response tasks to the edge, and low-latency tasks to the device side, and dynamically adjusts task allocation.

[0006] In this technical solution, the rapid collection and preliminary processing of basic data are achieved through the end-side devices, the complex computing tasks are deeply analyzed by the edge nodes, and the overall strategy optimization is completed by the large cloud model, thereby improving the overall effect while meeting the real-time and accuracy requirements. Collaborative reasoning of large and small models becomes the core strategy for improving performance. Basic feature extraction is completed through the lightweight model on the end-side, and refined reasoning is performed in combination with the large model. The above-mentioned method of this application can improve the intelligence level of ore mining and improve resource utilization.

[0007] The present invention is further configured such that: the terminal side collects and analyzes the ore data, including: Sensors and inference models are deployed on the end side. The physical signals of the ore are collected through the sensors, the physical signals are preprocessed according to the inference model, and preliminary control parameters are derived based on the preprocessed physical signals.

[0008] In this technical solution, the end side mainly collects and analyzes physical signals. Through sensors and inference models, the mechanical vibration signals, pressure signals, sound signals, image signals and environmental signals of the ore are collected and analyzed, and data preprocessing and feature extraction are performed to generate preliminary control parameters.

[0009] The present invention is further configured such that: the edge performs in-depth analysis on the data uploaded by the end side to generate an optimization control instruction, including: Integrate data uploaded by end-side devices and generate optimized control instructions based on historical data and environmental variables; Dynamically adjust multi-device load distribution, material conveying paths, and screening strategies.

[0010] In this technical solution, a medium-complexity model is deployed on the edge computing node to integrate and deeply analyze the data uploaded by the end-side devices, generate optimized control instructions, and implement data buffering and preprocessing.

[0011] The present invention is further configured such that the cloud performs an in-depth analysis of ore characteristics and production chain, including: Conduct in-depth analysis of the physical and chemical properties of ores and optimize mining strategies based on large cloud-based models; New model parameters are trained based on data uploaded by edge nodes, and edge and device-side models are periodically updated.

[0012] In this technical solution, a corresponding large model is deployed in the cloud, which can conduct in-depth analysis of ore characteristics and production chains, including ore characteristic prediction, mining strategy planning and periodic updating of model parameters.

[0013] The present invention is further configured as follows: the coordination mechanism includes: Tasks are allocated in layers, with global optimization tasks allocated to the cloud and local optimization tasks allocated to the edge. Tasks are also adjusted in real time to the device side. Adaptive task scheduling, dynamically adjusting task distribution based on network conditions and computing load; Incremental updates: incremental updates to edge and device models are performed through the cloud.

[0014] In this technical solution, the collaborative mechanism mainly includes task hierarchical allocation, adaptive task scheduling and incremental update mechanism, which incrementally updates the edge and end models through the cloud to ensure long-term adaptability in complex mining environments.

[0015] The present invention is further configured as follows: Step S2 includes: S21: A collaborative model between the device, edge, and cloud leverages ore characteristics and real-time feedback to dynamically adjust key parameters of production equipment. In S22, the cloud trains new model parameters based on historical data and current production conditions, optimizes ore mining strategies, and sends the optimized model parameters to edge and end devices to complete updates and iterations.

[0016] In this technical solution, step S2 mainly includes a real-time dynamic adjustment process and a layered collaborative update process. The real-time dynamic adjustment process mainly dynamically adjusts the parameters of the production equipment based on the physical and chemical properties of the ore and real-time feedback; the layered collaborative update can send the optimized parameters to the edge and end-side devices to achieve updates and iterations of the entire system.

[0017] The present invention is further configured such that: the end side, edge side and cloud side are connected to each other through a low-latency communication protocol.

[0018] In this technical solution, an efficient and low-latency communication protocol is used for data communication and transmission to ensure the stability and reliability of data transmission.

[0019] An ore mining control system adopts the above-mentioned ore mining control method, comprising: The terminal equipment module collects and analyzes ore data and generates preliminary control parameters; The edge computing module integrates and deeply analyzes the data uploaded by the edge side and generates optimized control instructions; Cloud processing module, which conducts in-depth analysis and long-term optimization of ore characteristics and production chains based on large models; The communication module supports low-latency communication protocols between the device side, edge side, and cloud side.

[0020] In this technical solution, the end-side device module, edge-side computing module and cloud-side processing module are connected in sequence and are all connected through corresponding communication modules. Through the collaborative mechanism of the above-mentioned end-side device module, edge-side computing module and cloud-side processing module, efficient data processing, accurate decision-making and dynamic control are achieved in the mining process, solving the problems of inefficiency, delay and lack of accuracy in the existing technology.

[0021] The present invention is further configured as follows: the terminal side device module includes: Sensor module, collects physical signals of ore in real time; The inference module pre-processes the collected physical signals and generates preliminary control parameters; The equipment control module receives control parameters and dynamically adjusts the operating status of the production equipment.

[0022] In this technical solution, the end-side device module mainly includes a sensor module, an inference module and a device control module. The sensor module is connected to the inference module, and the device control module is connected to the sensor module and the inference module respectively. The end-side device module can complete signal acquisition and analysis, data preprocessing and inference of preliminary control parameters.

[0023] The present invention is further configured such that: the edge computing module includes: The edge inference and integration module receives data uploaded by the end side, combines it with historical information for deep reasoning, and generates optimized device control instructions; The data synchronization and communication module completes efficient data exchange between the terminal and the cloud, while integrating data from multiple devices and uploading it to the cloud.

[0024] In this technical solution, the edge computing module mainly includes an edge reasoning and integration module and a data synchronization and communication module. The integration module and the data synchronization and communication module are connected to jointly realize the functions of the edge computing module.

[0025] The present invention can bring the following beneficial effects: 1. This invention improves the efficiency of data processing and model reasoning through a collaborative mechanism among the device, edge, and cloud. 2. This invention can combine large cloud models with edge reasoning to achieve high-precision control of the ore mining process; 3. The present invention reduces network transmission delay through hierarchical task allocation and is suitable for mining in remote mountainous areas; 4. The present invention can adjust strategies according to ore characteristics and environmental changes through adaptive update mechanism and collaborative optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1This is a schematic diagram of an ore mining control system of the present application.

[0027] Figure 2 It is a flow chart of an ore mining control method of the present application. DETAILED DESCRIPTION

[0028] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific implementation method described here is only an optimal embodiment of the present invention, which is only used to explain the present invention and does not limit the scope of protection of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0029] Example 1 This embodiment proposes a method for controlling ore mining. Figure 1 and Figure 2 , which mainly includes the following steps.

[0030] In step S1, data is collected on the end side, ore data is collected and analyzed; then, the data uploaded from the end side is integrated and deeply analyzed on the edge side to obtain optimized control instructions; finally, the ore characteristics and production chain are deeply analyzed and optimized in the long term on the cloud.

[0031] The content of step S1 is further expanded below.

[0032] The specific implementation process for collecting and analyzing ore data on the end side is: to achieve it by deploying corresponding sensors and inference models on the end side, further, to collect the physical signals of the ore through the above-mentioned sensors, to preprocess the physical signals through the above-mentioned inference model, and after the physical signals are preprocessed, to derive preliminary control parameters.

[0033] The above-mentioned preliminary control parameters mainly include the crusher instantaneous speed adjustment, the dynamic setting of the sieve opening and the hydraulic system pressure regulation.

[0034] In this technical solution, the end side mainly collects and analyzes physical signals. Through sensors and inference models, the mechanical vibration signals, pressure signals, sound signals, image signals and environmental signals of the ore are collected and analyzed, and data preprocessing and feature extraction are performed to generate preliminary control parameters.

[0035] The edge performs in-depth analysis of data uploaded from the device to generate optimized control instructions. This process involves integrating data uploaded by devices on the device side, combining it with historical data and environmental variables to generate optimized control instructions. This process then dynamically adjusts multi-device load distribution, material delivery routes, and screening strategies. Furthermore, the edge acts as a bridge between the device side and the cloud, buffering data and reducing network pressure.

[0036] In this technical solution, a medium-complexity model is deployed on the edge computing node to integrate and deeply analyze the data uploaded by the end-side devices, generate optimized control instructions, and implement data buffering and preprocessing.

[0037] The specific implementation process for in-depth analysis of ore characteristics and production chains in the cloud is as follows: a corresponding large data model is deployed in the cloud, based on which the physical and chemical properties of the ore can be deeply analyzed and the mining strategy optimized; then new model parameters are trained based on the data uploaded by the edge nodes, and the edge and end models are periodically updated; long-term production chain planning is carried out to maximize resource utilization efficiency.

[0038] In this technical solution, a corresponding large model is deployed in the cloud, which can conduct in-depth analysis of ore characteristics and production chains, including ore characteristic prediction, mining strategy planning and periodic updating of model parameters.

[0039] Step S2: Adjust the production equipment parameters according to the ore characteristics and real-time feedback. The cloud can train and optimize the large model based on the production data and update the optimized parameters to the edge and end sides.

[0040] As for step S2, it mainly includes a real-time dynamic adjustment process and a hierarchical collaborative update process.

[0041] For the real-time dynamic adjustment process, that is, step S21, the collaborative model between the end-side, edge-side and cloud-side can dynamically adjust the key parameters of the production equipment based on the ore characteristics and real-time feedback.

[0042] For the layered collaborative update process, that is, step S22, the cloud can train new model parameters based on historical data and current production conditions, and optimize the ore mining strategy. The optimized model parameters are then sent to the guide edge and end-side devices to achieve updates and iterations of the entire system.

[0043] In this technical solution, step S2 mainly includes a real-time dynamic adjustment process and a layered collaborative update process. The real-time dynamic adjustment process mainly dynamically adjusts the parameters of the production equipment based on the physical and chemical properties of the ore and real-time feedback; the layered collaborative update can send the optimized parameters to the edge and end-side devices to achieve updates and iterations of the entire system.

[0044] In step S3, through the collaborative mechanism, specifically, global optimization tasks are assigned to the cloud, fast response tasks are assigned to the edge, and low latency tasks are assigned to the device side, dynamically adjusting task distribution and optimizing response time.

[0045] As for the coordination mechanism in step S3, in more detail, it mainly includes task hierarchical allocation, adaptive task scheduling and incremental update mechanism.

[0046] For task hierarchical allocation, the corresponding tasks are sent to the corresponding layers. Specifically, the global optimization tasks are assigned to the cloud, the fast response tasks are assigned to the edge, and the low-latency tasks are assigned to the device side.

[0047] For adaptive task scheduling, it mainly adjusts task distribution according to network conditions and computing load dynamics.

[0048] As for the incremental update mechanism, it mainly performs incremental updates on the end-side and edge-side models based on the cloud to ensure the long-term effectiveness of the model and its long-term adaptability in complex mineral environments.

[0049] In this technical solution, the collaborative mechanism mainly includes task hierarchical allocation, adaptive task scheduling and incremental update mechanism, which incrementally updates the edge and end models through the cloud to ensure long-term adaptability in complex mining environments.

[0050] During the above steps S1 to S3, the device side, edge side, and cloud side specifically interact with each other through a low-latency communication protocol to ensure the reliability of task distribution and data transmission, and optimize the response time through a layered distribution mechanism.

[0051] In this technical solution, an efficient and low-latency communication protocol is used for data communication and transmission to ensure the stability and reliability of data transmission.

[0052] Based on the above-mentioned ore mining control method, this embodiment also proposes an ore mining control system, referring to Figure 1 and Figure 2 It mainly includes an end-side device module, an edge computing module, a cloud processing module and a communication module. The end-side device module, the edge computing module and the cloud processing module are connected in sequence and are all connected through corresponding communication modules.

[0053] For the end-side equipment module, its main functions are: collecting and analyzing ore data, and finally obtaining preliminary control parameters; for more detailed functions of the end-side equipment module, please refer to the description of step S1 of the above-mentioned ore mining control method.

[0054] For the edge computing module, its main functions are: integrating the data uploaded by the end side, and conducting in-depth analysis on the integrated data, and finally obtaining optimized control instructions; for more detailed functions of the edge computing module, please refer to the description of step S1 of the above-mentioned ore mining control method.

[0055] The cloud processing module mainly implements the following functions: in-depth analysis and long-term optimization of ore characteristics and production chains based on the large model it establishes. For more detailed functions of the cloud processing module, please refer to the description of steps S1 to S3 of the above-mentioned ore mining control method.

[0056] The communication module primarily supports low-latency communication protocols between devices, edge devices, and the cloud. It ensures the reliability of task distribution and data transmission, optimizing response time through a layered distribution mechanism. More specifically, the communication module utilizes a low-latency, high-reliability layered distribution mechanism to ensure efficient data transmission between devices, edge devices, and the cloud.

[0057] In this technical solution, through the collaborative mechanism of the above-mentioned end-side device module, edge computing module and cloud processing module, efficient data processing, accurate decision-making and dynamic control are achieved in the mining process, solving the problems of inefficiency, delay and lack of accuracy in the existing technology.

[0058] As for the end-side device module, to be more specific, it mainly includes a sensor module, an inference module and a device control module. The sensor module is connected to the inference module, and the device control module is connected to the sensor module and the inference module respectively.

[0059] The main function of the sensor module is to collect the physical signals of the ore in real time.

[0060] For the inference module, its main function is to preprocess the collected physical signals and, after preprocessing, infer preliminary control parameters.

[0061] The main functions of the equipment control module are: accepting the above control parameters and dynamically adjusting the operating status of the production equipment.

[0062] In this technical solution, the end-side device module mainly includes a sensor module, an inference module and a device control module. The sensor module is connected to the inference module, and the device control module is connected to the sensor module and the inference module respectively. The end-side device module can complete signal acquisition and analysis, data preprocessing and inference of preliminary control parameters.

[0063] In this embodiment, the physical signals collected by the sensor module include but are not limited to vibration, pressure, sound, image, and environmental signals. Sensors include but are not limited to vibration sensors, pressure sensors, sound sensors, image acquisition devices, and environmental monitoring sensors.

[0064] In this embodiment, the preprocessing process in the inference module includes but is not limited to denoising, enhancement, and feature extraction.

[0065] In this embodiment, the equipment control module can adjust the crusher rotation speed, sieve opening size and hydraulic pressure.

[0066] For the edge computing module, it mainly includes an edge reasoning and integration module and a data synchronization and communication module. The edge reasoning and integration module and the data synchronization and communication module are connected.

[0067] The edge inference and integration module receives and integrates data uploaded from the device side, then performs deep inference based on historical information to ultimately generate optimized device control instructions. The data synchronization and communication module can realize the following functions: it can realize efficient data exchange between the terminal and the cloud, and integrate data from multiple devices and upload it to the cloud.

[0068] In this technical solution, the edge computing module mainly includes an edge reasoning and integration module and a data synchronization and communication module. The integration module and the data synchronization and communication module are connected to jointly realize the functions of the edge computing module.

[0069] The edge computing module achieves efficient resource utilization and task optimization by dynamically adjusting the load distribution of equipment and the material transportation path.

[0070] In addition, the cloud processing module mainly includes a large model training and optimization module and a collaborative control module. The main functions of the large model training and optimization module are: generating global mining strategies and optimizing resource allocation through large-scale data analysis; the main functions of the collaborative control module are: responsible for the optimization and distribution of model parameters of the entire system, and periodically updating the edge and end models.

[0071] The cloud module optimizes the operating parameters of ore mining equipment by analyzing historical data, reducing energy consumption and minimizing ore loss.

[0072] Example 2 Based on Example 1, an implementation method of an ore mining control method is proposed.

[0073] The process of collecting and analyzing ore data on the end side in step S1 is described in detail here.

[0074] Real-time signal collection and analysis: Lightweight sensors and inference models (with approximately 0.5B parameters) are deployed on end-side devices such as ore crushers to collect physical signals of the ore in real time: for example, mechanical vibration signals (to detect the operating status of the crusher and the crushing characteristics of the ore), pressure sensor signals (to reflect the hardness of the ore and the machine workload), sound signals (to analyze the flow state of the ore in the equipment), image signals (using cameras to monitor the particle size, shape and material flow of the ore), and environmental signals (temperature, humidity, and other external conditions).

[0075] After completing the above-mentioned signal acquisition and analysis, the data is preprocessed. In this embodiment, the preprocessing includes eliminating noise and enhancing signals; extracting key features such as frequency, amplitude, texture, and image contour; and inferring preliminary control parameters, such as instantaneous speed adjustment of the crusher, dynamic setting of the sieve opening size, and pressure regulation of the hydraulic system.

[0076] The process of performing in-depth analysis on the data uploaded by the edge-to-end side to generate optimized control instructions in step S1 is described in detail here.

[0077] A model of medium complexity (usually 7B or 13B parameters) is deployed on the edge computing node (edge ​​server) to integrate and deeply analyze the data uploaded by the edge devices: historical data and environmental variables are combined to generate optimized control instructions. For example, the hardness distribution of ore is analyzed through vibration signals to dynamically adjust the load distribution of multiple crushers; the material transportation path is adjusted based on the screening results to avoid resource waste.

[0078] The process of in-depth analysis of ore characteristics and production chain in the cloud in step S1 is described in detail here.

[0079] Deploying a complete large model (70B or more parameter volume) in the cloud enables in-depth analysis of ore characteristics and long-term optimization of the production chain; ore characteristic analysis and long-term planning decisions for mining strategies; for example, predicting the optimal operating parameters of equipment based on the physical and chemical properties of the ore; adjusting crushing strategies to reduce energy consumption based on historical mining data; optimizing screening and conveying systems to minimize ore loss; training new model parameters based on data uploaded by edge nodes, and periodically updating edge and end-side models to improve their performance.

[0080] As for step S2, it mainly includes real-time dynamic adjustment and hierarchical collaborative updating processes.

[0081] The process of real-time dynamic adjustment is described in detail here.

[0082] The end-side-edge-cloud collaborative model dynamically adjusts key parameters of production equipment based on ore characteristics and real-time feedback, including: adjusting the crusher speed, optimizing the crusher speed in real time according to the ore hardness, particle size and equipment load to ensure crushing efficiency and equipment life; regulating the hydraulic system pressure, dynamically adjusting the hydraulic pressure according to the ore's compressive properties and workload status to avoid equipment overload or energy waste.

[0083] The coordination mechanism in step S3 is described in detail here.

[0084] The collaborative mechanism mainly includes hierarchical task allocation, adaptive task scheduling and incremental update mechanism.

[0085] During the implementation of hierarchical task allocation, tasks that require large amounts of computation and global optimization, such as mining strategy optimization and long-cycle equipment planning, are assigned to the cloud; tasks that require fast response, such as multi-device load distribution and local parameter optimization, are assigned to edge devices; low-latency real-time tasks, such as device status adjustment, are assigned to end-side devices.

[0086] For adaptive task scheduling, it can dynamically adjust task distribution based on current network conditions and computing load, optimize bandwidth utilization and reduce response time.

[0087] As for the incremental update mechanism, it is specifically as follows: the cloud optimizes the incremental update content of the edge and end-side models to improve performance; adjusts the optimization strategy through real-time feedback to ensure long-term adaptability in complex mineral environments.

[0088] Example 3 Based on the above-mentioned embodiment 1 or embodiment 2, this embodiment proposes a more specific implementation method, which mainly involves the processing of ores with varying hardness.

[0089] First, an end-side response is performed: when the vibration sensor detects an increase in the hardness of the ore, the end-side lightweight model quickly analyzes the signal characteristics (such as changes in vibration frequency and amplitude) and identifies the characteristics of hard ore. To avoid equipment overload, the crusher speed is immediately reduced and the crushing time is extended, thereby ensuring that the ore is fully crushed and reducing equipment wear.

[0090] Furthermore, edge optimization is performed: edge nodes aggregate data from multiple devices (such as crusher operating status, material flow, and hardness distribution) and, combined with the current ore processing load, generate optimization solutions. For example, they dynamically adjust the operating parameters or load distribution of other devices, shifting some processing tasks to backup devices to prevent downtime due to overload. Simultaneously, edge nodes provide feedback on optimization strategies to edge devices to improve efficiency.

[0091] Finally, cloud-based strategy adjustments are made: the cloud-based model performs trend analysis on multiple batches of ore hardness data to predict the potential demand for centralized processing of hard ores in the future. Based on the analysis results, the cloud-based large model determines the optimal mining route and equipment selection, and optimizes the entire system by distributing parameters.

[0092] Finally, it should be noted that the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions to the technical solutions of the present invention do not depart from the spirit and scope of the technical solutions of the present invention and are intended to be encompassed by the claims of the present invention.

Claims

1. A method for controlling ore mining, characterized in that: The following steps are involved: S1: The end side collects and analyzes ore data; the edge side performs in-depth analysis on the data uploaded by the end side to generate optimization control instructions; the cloud side performs in-depth analysis and optimization on ore characteristics and production chains; S2, adjusts production equipment parameters according to ore characteristics and real-time feedback. The cloud trains and optimizes the large model based on production data, and updates the optimized parameters to the edge and end sides; S3, through a collaborative mechanism, allocates global optimization tasks to the cloud, fast response tasks to the edge, low latency tasks to the device side, and dynamically adjusts task allocation.

2. A method for controlling ore mining according to claim 1, characterized in that: The end side collects and analyzes the data of the ore, including: Sensors and inference models are deployed on the end side. Physical signals of ore are collected through sensors. The physical signals are preprocessed according to the inference model. Preliminary control parameters are derived based on the preprocessed physical signals.

3. The ore mining control method according to claim 1, characterized in that: The edge performs in-depth analysis on the data uploaded by the end side to generate optimization control instructions, including: Integrate the data uploaded by the end-side devices, and generate optimized control instructions based on historical data and environmental variables; Dynamically adjust multi-device load distribution, material conveying paths and screening strategies.

4. The ore mining control method according to claim 1, characterized in that: The cloud provides in-depth analysis of ore characteristics and production chains, including: A large model deployed in the cloud is used to conduct in-depth analysis of the physical and chemical properties of ores and optimize mining strategies; New model parameters are trained based on the data uploaded by edge nodes, and edge and device models are updated periodically.

5. A method for controlling ore mining according to any one of claims 1 to 4, characterized in that: The coordination mechanism includes: Tasks are allocated in layers, with global optimization tasks allocated to the cloud and local optimization tasks allocated to the edge. Tasks are allocated to the device in real time. Adaptive task scheduling, dynamically adjusting task distribution based on network conditions and computing load; Incremental update: incremental update of edge and device models through the cloud.

6. A method for controlling ore mining according to claim 5, characterized in that: The step S2 comprises: S21, a collaborative model between the end-side, edge-side, and cloud uses ore characteristics and real-time feedback to dynamically adjust key parameters of production equipment; S22, the cloud trains new model parameters based on historical data and current production conditions, optimizes ore mining strategies, and sends the optimized model parameters to edge and end devices to complete updates and iterations.

7. A method for controlling ore mining according to claim 6, characterized in that: The end side, edge side and cloud side are connected to each other through a low-latency communication protocol.

8. A mineral mining control system, using a mineral mining control method according to any one of claims 1 to 7, characterized in that: include: The terminal equipment module collects and analyzes the ore data and generates preliminary control parameters; The edge computing module integrates and deeply analyzes the data uploaded by the end side and generates optimized control instructions; Cloud processing module for in-depth analysis and long-term optimization of ore characteristics and production chains based on big models; The communication module supports low-latency communication protocols between the client, edge, and cloud.

9. A method for controlling ore mining according to claim 8, characterized in that: The terminal side device module includes: Sensor module, collects physical signals of ore in real time; The inference module pre-processes the collected physical signals and generates preliminary control parameters; The equipment control module receives control parameters and dynamically adjusts the operating status of the production equipment.

10. A method for controlling ore mining according to claim 8 or 9, characterized in that: The edge computing module includes: The edge reasoning and integration module receives data uploaded by the end side, performs deep reasoning based on historical information, and generates optimized device control instructions. The data synchronization and communication module completes efficient data exchange between the terminal and the cloud, while integrating data from multiple devices and uploading it to the cloud.

Images