Data monitoring system in uranium mine production process

Abstract

The invention discloses a data monitoring system in a uranium mine production process, relates to the technical field of data processing, and mainly aims to solve the problem of poor data monitoring effectiveness in the existing uranium mine production process. Comprising a visualization subsystem, a data processing subsystem and a data monitoring subsystem, and the visualization subsystem is used for constructing a semantic three-dimensional model of the uranium mine production process based on uranium mine production components and visually displaying multi-source data based on the semantic three-dimensional model. The semantic three-dimensional model comprises geometric elements, texture elements, semantic elements and relation elements in the production process; the data processing subsystem is used for collecting multi-source data in real time, marking and fusing the multi-source data and sending the multi-source data to the data monitoring subsystem; and the data monitoring subsystem is used for performing fault diagnosis on the marked and fused multi-source data based on a transfer learning algorithm so as to display an obtained fault diagnosis result in the visualization subsystem.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a data monitoring system for uranium mine production processes. Background Technology

[0002] With the application of digitalization in industrial production, more and more industrial production is adopting big data processing methods for production monitoring. Among them, the uranium mine production process includes the injection process, the dosing process, and the leaching solution treatment process. In order to ensure the safety of industrial production, it is necessary to conduct process control such as analysis and monitoring of the above production processes, remote equipment control, daily inspection, and data statistical analysis.

[0003] Currently, monitoring of uranium mine production processes typically employs automated monitoring systems to collect data from different production stages, using corresponding numerical analysis and manual inspection methods for each stage. However, due to the complexity of uranium hydrometallurgical processes, with numerous production stages, equipment, and pipelines, systems such as production automation, laboratory analysis, video surveillance, production management, and environmental monitoring operate independently, data cannot be shared, and decision-making relies on manual data aggregation, significantly impacting the effectiveness of monitoring. Furthermore, reliance on manual on-site inspections of pump stations, pipelines, and instruments creates blind spots and lags in response, posing significant safety hazards in high-risk areas, thus failing to meet the effective data monitoring needs of uranium mine production processes. Summary of the Invention

[0004] In view of this, this application provides a data monitoring system for the uranium mine production process, the main purpose of which is to solve the problem of poor data monitoring effectiveness in the existing uranium mine production process.

[0005] According to one aspect of this application, a data monitoring system for uranium mine production processes is provided, comprising: a visualization subsystem, a data processing subsystem, and a data monitoring subsystem. The visualization subsystem is used to construct a semantic 3D model of the uranium mine production process based on uranium mine production components, and to visualize multi-source data based on the semantic 3D model. The semantic 3D model includes geometric elements, texture elements, semantic elements, and relational elements in the production process. The data processing subsystem is used to collect the multi-source data in real time, and to label and fuse the multi-source data before sending it to the data monitoring subsystem. The data monitoring subsystem is used to perform fault diagnosis on the multi-source data after label fusion based on the transfer learning algorithm, and to display the obtained fault diagnosis results in the visualization subsystem.

[0006] Furthermore, the semantic 3D model includes a geometric sub-model and a process flow sub-model; The visualization subsystem is specifically used to add semantic information and rule function information to the uranium mine production components on the geometric sub-model, determine the process flow information of the uranium mine production process, and adjust the process flow sub-model according to the process flow information. The visualization subsystem is further configured to combine the geometric sub-model with the process flow sub-model, and mark geometric elements, texture elements, semantic elements and relational elements according to the semantic information, rule function information and process flow information to generate the semantic 3D model.

[0007] Furthermore, the data processing subsystem is specifically used to acquire multi-source data collected by different multi-source devices through an integrated interface, wherein the multi-source data includes at least one of image data, signal data, numerical data, and text data; The data processing subsystem is further used to extract features from the multi-source data, label the extracted data features, and store them in the database.

[0008] Furthermore, the data processing subsystem is also used to retrieve data features from the database that match the tags carried in the data retrieval instruction after receiving a data retrieval instruction from the data monitoring subsystem; The data processing subsystem is also used to perform feature fusion on the data features and feed them back to the data monitoring subsystem.

[0009] Furthermore, the data monitoring subsystem is also used to determine the target marker and send a data retrieval command in response to the detection command; The data monitoring subsystem is also used to identify the data features of the target labeled as an image based on a visual recognition model to obtain visual diagnostic results; The data monitoring subsystem is also used to identify the data features of the target labeled as numerical or text based on the environmental identification model, and obtain environmental diagnosis results.

[0010] Furthermore, the data monitoring subsystem is also used to associate and bind the three-dimensional coordinates of the semantic three-dimensional model with the data time series library, and to send an early warning command to the visualization subsystem when data points in the data time series library are diagnosed as abnormal. The visualization subsystem is also used to respond to the warning command by generating a warning area using the three-dimensional coordinates and visually displaying the warning area.

[0011] Furthermore, the visualization subsystem is also used to obtain the multi-source data from the data processing subsystem, generate a production process consumption curve corresponding to the multi-source data, and perform time-series visualization early warning on the production process consumption curve through a consumption threshold.

[0012] Furthermore, the data monitoring subsystem is also used to determine the inspection object based on distance and online status when the fault diagnosis result indicates that a fault has occurred, and generate an inspection path; and send the inspection path to the inspection object to perform offline inspection.

[0013] Furthermore, the data monitoring subsystem is also used to generate an emergency decision-making strategy based on the fault diagnosis results and send it to the visualization subsystem to output the emergency decision-making strategy.

[0014] Furthermore, the data monitoring subsystem is also used to acquire the multi-source data, generate statistical results of the multi-source data according to a preset time interval, and output them.

[0015] By employing the above technical solutions, the technical solutions provided in the embodiments of this application have at least the following advantages: This application provides a data monitoring system for uranium mine production processes, including a visualization subsystem, a data processing subsystem, and a data monitoring subsystem. The visualization subsystem is used to construct a semantic 3D model of the uranium mine production process based on uranium mine production components, and to visualize multi-source data based on the semantic 3D model. The semantic 3D model includes geometric elements, texture elements, semantic elements, and relational elements related to the production process. The data processing subsystem is used to collect the multi-source data in real time, and to perform labeling and fusion on the multi-source data before sending it to the data monitoring subsystem. The data monitoring subsystem is used to perform fault diagnosis on the labeled and fused multi-source data based on a transfer learning algorithm, and to display the obtained fault diagnosis results in the visualization subsystem. This system achieves full intelligentization of the uranium mine production process, greatly improves data integration and monitoring effectiveness, reduces material and labor costs, and realizes integrated comprehensive management and control of online monitoring, remote equipment control, and data statistical analysis, thereby improving production safety and meeting the effective data monitoring needs of uranium mine production processes.

[0016] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This paper illustrates a schematic diagram of a data monitoring system structure for a uranium mine production process, as provided in an embodiment of this application. Figure 2 This illustration shows a hierarchical structure diagram of a data monitoring system for uranium mine production provided in an embodiment of this application. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0019] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] This application provides a data monitoring system for the uranium mine production process, such as... Figure 1 As shown, the system includes: a visualization subsystem 101, a data processing subsystem 102, and a data monitoring subsystem 103. The visualization subsystem is used to construct a semantic three-dimensional model of the uranium mine production process based on uranium mine production components, and to visualize and display multi-source data based on the semantic three-dimensional model. The data processing subsystem is used to collect the multi-source data in real time, and to label and fuse the multi-source data before sending it to the data monitoring subsystem. The data monitoring subsystem is used to perform fault diagnosis on the multi-source data after label fusion based on the transfer learning algorithm, and to display the obtained fault diagnosis results in the visualization subsystem.

[0021] In this embodiment, the visualization subsystem acts as the main body for generating the semantic 3D model. It constructs a 3D model of the uranium mine production process, providing full visualization of the entire process. The uranium mine production components are constructed from views within 3D software that builds the 3D model. These components can be selected based on the equipment and operational procedures required in the uranium mine production process; no specific limitation is placed on the 3D software. The semantic 3D model includes geometric elements, texture elements, semantic elements, and relational elements. Geometric elements represent the geometric shapes of various equipment or processes during production. Texture elements represent the visualized texture content of each piece of equipment or process. Semantic elements represent the semantic content requiring textual description in each piece of equipment or process. Relational elements represent the logical or functional relationships between different pieces of equipment or processes. Furthermore, the multi-source data consists of data collected from different sensing devices, including but not limited to image data, signal data, numerical data, and text data, which the data processing subsystem uses to label and fuse the multi-source data before sending it to the data monitoring subsystem. When the data monitoring subsystem receives labeled and fused multi-source data, it can perform fault diagnosis on the labeled and fused multi-source data based on a transfer learning algorithm, and display the obtained fault diagnosis results in the visualization subsystem. Specifically, the data monitoring subsystem can pre-train targeted transfer learning algorithms for different fault diagnoses to meet different fault diagnosis needs; this application embodiment does not impose specific limitations.

[0022] In another embodiment of this application, for further definition and explanation, The visualization subsystem is specifically used to add semantic information and rule function information to the uranium mine production components on the geometric sub-model, determine the process flow information of the uranium mine production process, and adjust the process flow sub-model according to the process flow information. The visualization subsystem is further configured to combine the geometric sub-model with the process flow sub-model, and mark geometric elements, texture elements, semantic elements and relational elements according to the semantic information, rule function information and process flow information to generate the semantic 3D model.

[0023] To achieve full visualization of uranium mine production, the visualization subsystem, when constructing a semantic 3D model using 3D software, first adds semantic information and rule-based functional information to all components based on the geometric sub-model. At this point, the geometric sub-model is either a combination of selected uranium mine production components or a default geometric sub-model in the 3D software. Further, the basic components of the combined geometric sub-model are filtered to determine the uranium mine production components. The semantic information may include, but is not limited to, the component's equipment name, size, and geographical location. The rule-based functional information characterizes the component's attributes, business information, and functions during uranium mine production, and may include, but is not limited to, static attributes of facilities and equipment, spatial relationships with upstream and downstream equipment, and control logic relationships. This application's embodiments do not impose specific limitations on these aspects. Simultaneously, the process flow information of the uranium mine production process is determined, and the process flow sub-model is adjusted according to the process flow information. At this time, the process flow information includes the entire production process of the uranium mine, which can be obtained by manual uploading. Then, the default or initial process flow sub-model in the 3D software is adjusted. For example, fluid-related data is obtained, and the lightweight fluid dynamics model is adjusted to obtain a 3D process flow sub-model suitable for the embodiments of this application.

[0024] It should be noted that, in order for the semantic 3D model to display the content of both the geometric sub-model and the process flow sub-model, the visualization subsystem combines the geometric sub-model and the process flow sub-model, and marks the geometric elements, texture elements, semantic elements, and relational elements according to the semantic information, rule function information, and process flow information to generate the semantic 3D model. At this time, the process flow information may include, but is not limited to, all process contents in the uranium ore production process, which may be manually set, and then each element is marked. The marking may be based on manual marking or based on identifiers or types. This application embodiment does not make specific limitations.

[0025] In some embodiments, the visualization subsystem adds static attributes of facilities and equipment to each component, associates upstream and downstream equipment spaces and control logic relationships, upgrading the model from a shape to a functional entity. The dynamic attributes of the model are driven by real-time data, and the model state can be updated synchronously, realizing digital mirroring. In addition, the process flow sub-model can be an integrated computational fluid dynamics lightweight model, realizing real-time rendering based on physical laws. It can simulate and visualize flow routes, concentration distributions, etc. in real time based on real-time flow rate, concentration, permeability, etc. The modeling covers millimeter-level precision models of general appearance, buildings, process equipment pipelines, electrical automatic control instruments, etc. It integrates geological body models (ore layers, aquifers), dynamically renders the distribution of ore-bearing layers, liquid diffusion range, etc., and constitutes virtual production scenarios for different process sections.

[0026] In some embodiments, the visualization subsystem is configured with a 2D / 3D fusion interface, which can realize the visualization and monitoring of the production process. Clicking on any device in the virtual scene allows for remote control, enabling data-driven simulation of processes such as hydrometallurgical ion exchange, material transport, and injection, as well as real-time updates of equipment status and visualization of real-time alarm information.

[0027] In another embodiment of this application, for further definition and explanation, The data processing subsystem is specifically used to acquire multi-source data collected by different multi-source devices through an integrated interface. The data processing subsystem is further used to extract features from the multi-source data, label the extracted data features, and store them in the database.

[0028] To ensure data validity, the data processing subsystem can acquire multi-source data collected from various devices (such as electrical control instruments, temperature gauges, barometers, cameras, etc.) via an integrated interface API. This multi-source data includes at least one of the following: image data, signal data, numerical data, and text data. Simultaneously, the data processing subsystem extracts features from the multi-source data, labels the extracted data based on these features, and stores them in the database. Feature extraction can be performed using machine learning algorithms such as autoencoders and convolutional neural networks. During labeling, the extracted data features can be categorized, sorted, and labeled for storage in the corresponding database of the data processing subsystem.

[0029] In some embodiments, when the data processing subsystem acquires multi-source data collected from different multi-source devices through an integration interface, it can integrate and process multi-source data such as electrical, automatic control, analysis, and video data through an industrial data interface. This data is stored in a unified database and can be correlated with events. A semantic 3D model serves as the hub for data integration, with each model component identified by a unique ID, becoming the core index for aggregating all related data. Simultaneously, a streaming computing engine can be used to perform online feature extraction and mine condition identification on the acquired real-time data. Furthermore, an API based on model component IDs can be provided to support intelligent applications, facilitating the acquisition of full-dimensional data from any device across systems and time periods.

[0030] In another embodiment of this application, for further definition and explanation, The data processing subsystem is also used to retrieve data features that match the tags carried in the data retrieval instruction from the database when it receives a data retrieval instruction from the data monitoring subsystem. The data processing subsystem is also used to perform feature fusion on the data features and feed them back to the data monitoring subsystem.

[0031] To enable the data subsystem to serve as data support in uranium mine production monitoring, the data processing subsystem, upon receiving a data retrieval command from the data monitoring subsystem, retrieves data features from the database that match the tags carried in the data retrieval command. This data can include, but is not limited to, historical or real-time data. Furthermore, the data processing subsystem can perform feature fusion on the data features and feed it back to the data monitoring subsystem. The feature fusion method can employ adaptive weighted fusion or multimodal fusion; this embodiment does not impose specific limitations.

[0032] In another embodiment of this application, for further definition and explanation, The data monitoring subsystem is also used to respond to detection commands, determine target markers, and send data retrieval commands. The data monitoring subsystem is also used to identify the data features of the target labeled as an image based on a visual recognition model to obtain visual diagnostic results; The data monitoring subsystem is also used to identify the data features of the target labeled as numerical or text based on the environmental identification model, and obtain environmental diagnosis results.

[0033] To meet diverse fault diagnosis needs, the data monitoring subsystem can also identify data features labeled as images based on a visual recognition model to obtain visual diagnostic results. This visual recognition model can be a graph convolutional neural network model, pre-trained using fault image samples for visual diagnosis; however, this embodiment does not impose specific limitations on this model. Simultaneously, it can also identify data features labeled as numerical or text based on an environmental recognition model to obtain environmental diagnostic results. This environmental recognition model can be a deep learning algorithm or a data prediction model such as a support vector machine, pre-trained using fault environment samples for environmental diagnosis; again, this embodiment does not impose specific limitations on this model.

[0034] In some embodiments, the data processing subsystem can also preset inspection routes, automatically acquire data such as video along the route, sensor data, and pipeline leakage status, and integrate an AI visual recognition model to automatically identify instrument data, leaks, and illegal intrusions. Simultaneously, based on real-time equipment health status, historical faults, and environmental risk assessments, it can dynamically generate optimal inspection routes through path planning algorithms, prioritizing high-risk points to achieve automatic inspection of virtual production scenarios. Upon detecting anomalies, it can automatically notify nearby personnel. Furthermore, AR glasses can overlay virtual information such as 3D fault models, maintenance manuals, and internal structures onto the real-world view, guiding efficient and accurate on-site handling. Additionally, by associating 3D coordinates with a unified time-series library, an interactive mode can be implemented to retrieve real-time data by clicking on the model. An event-driven architecture is established; when an anomaly occurs at a data point in the time-series library, the system automatically pushes an alarm event to the 3D scene to automatically switch perspectives, locate faulty equipment, and draw custom warning areas in the 3D scene. Illegal intrusion or sensor data exceeding limits can automatically trigger predefined actions, achieving unmanned automatic early warning.

[0035] In another embodiment of this application, for further definition and explanation, The data monitoring subsystem is also used to associate and bind the three-dimensional coordinates of the semantic three-dimensional model with the data time series library, and to send an early warning command to the visualization subsystem when data points in the data time series library are diagnosed as abnormal. The visualization subsystem is also used to respond to the warning command by generating a warning area using the three-dimensional coordinates and visually displaying the warning area.

[0036] To meet the need for precise monitoring of the entire uranium mining process, the data monitoring subsystem can associate and bind the 3D coordinates of a semantic 3D model with a time-series data library, ensuring that each 3D coordinate corresponds to data collected in a time sequence. Simultaneously, after the data monitoring subsystem determines diagnostic results based on various diagnostic models, it marks the diagnostic data against the data in the time-series data library as abnormal. Therefore, when an anomaly is diagnosed in a data point in the time-series data library, an early warning command can be sent to the visualization subsystem for alerting.

[0037] In another embodiment of this application, for further definition and explanation, The visualization subsystem is also used to obtain the multi-source data from the data processing subsystem, generate a production process consumption curve corresponding to the multi-source data, and perform time-series visualization early warning on the production process consumption curve through a consumption threshold.

[0038] To meet the real-time monitoring requirements of multi-source data, the visualization subsystem acquires multi-source data from the data processing subsystem and generates production process consumption curves corresponding to the multi-source data. These production process consumption curves represent the dynamics of different production consumption patterns represented by the multi-source data during the production process. The curves can be generated based on, but are not limited to, directly according to data fluctuations, or after relevant calculations; this application does not impose specific limitations. Furthermore, a threshold judgment method can be used to determine whether the production process consumption curves are abnormal. Consumption thresholds can be configured based on different curves, and abnormal curves can be alerted according to a time sequence; this application does not impose specific limitations.

[0039] In another embodiment of this application, for further definition and explanation, The data monitoring subsystem is also used to determine the inspection object based on distance and online status when the fault diagnosis result indicates that a fault has occurred, and generate an inspection path; and send the inspection path to the inspection object to perform offline inspection.

[0040] To improve the efficiency of inspections in uranium mine production processes, the fault diagnosis results obtained by the data monitoring subsystem determine the distance requiring inspection based on the faulty equipment or production process, as well as which inspection personnel are available for inspection (indicating online status). This allows for the identification of inspection targets (specific inspection personnel) based on distance and online status, and the generation of inspection paths according to the real-time distance of the inspection personnel and the fault distance. The data monitoring subsystem then sends the inspection path to the inspection targets for offline inspection.

[0041] In another embodiment of this application, for further definition and explanation, The data monitoring subsystem is also used to generate emergency decision-making strategies based on the fault diagnosis results and send them to the visualization subsystem to output the emergency decision-making strategies.

[0042] To enable the data monitoring system to provide effective solutions for staff, the data monitoring subsystem can generate emergency decision-making strategies based on fault diagnosis results and send them to the visualization subsystem to output the emergency decision-making strategies. The generation of emergency decision-making strategies based on visual diagnostic results, environmental diagnostic results, or other equipment diagnostic results can be achieved through methods including, but not limited to, comparing different diagnostic results with preset emergency decision-making strategies, and real-time manual loading, etc., which are not specifically limited in this application embodiment.

[0043] In another embodiment of this application, for further definition and explanation, The data monitoring subsystem is also used to acquire the multi-source data, generate statistical results of the multi-source data according to a preset time interval, and output them.

[0044] To facilitate data visualization and analysis, the data monitoring subsystem can also generate and output statistical results of multi-source data after acquiring multi-source data at preset time intervals (such as 1 month or 2 months). These statistical results include, but are not limited to, average statistics and total statistics; however, this embodiment does not impose specific limitations on them.

[0045] In some embodiments, when the data processing subsystem performs fault diagnosis, the online diagnosis and predictive maintenance of pump equipment integrates temperature, vibration, and electrical parameters to construct a fault diagnosis prediction model based on transfer learning, such as impeller wear and bearing failure. Using transfer learning technology, the model is pre-trained on publicly available vibration and temperature datasets for large-scale machinery equipment, and then fine-tuned using limited field data from uranium mines. Simultaneously, it integrates start-up and shutdown data and load rates, using causal inference to analyze the causes of faults, escalating from predicting faults to diagnosing root causes. Production data statistical analysis can generate curves for material consumption and energy consumption statistics in real time. Online leakage monitoring data is superimposed on the corresponding locations of the three-dimensional model of the wastewater treatment evaporation tank, with the monitoring data originating from the bottom of the tank. The distributed fiber optic temperature and leakage monitoring sensor continuously senses temperature changes at all points on the bottom of the pool in real time. It accurately locates leaks, triggers alarms and emergency plans, and automatically activates a simulation model of solute transport in the leaking liquid. This model dynamically simulates the future diffusion path and impact range of pollutants in a 3D scene, providing precise and scientific decision-making support for emergency response and achieving visualized online control of safety and environmental protection. In case of equipment failure or abnormal parameters, it automatically locates the fault point in the 3D scene, displays a real-time video feed, sends an interlock shutdown command, and pushes alarm information and handling plans. Intelligent reports allow for flexible configuration of daily, monthly, and yearly reports on self-control, analysis, and management, enabling multi-dimensional comparison, analysis, statistics, and output of production data.

[0046] In some embodiments, such as Figure 2As shown, the data monitoring system in the uranium mine production process, as a comprehensive management and control platform 1, can be composed of a 3D visualization engine layer 2 as a visualization subsystem, a data fusion layer 3 as a data processing subsystem, and an intelligent application layer 4 as a data monitoring subsystem. Among them, the 3D visualization engine layer 2 provides 3D models and dynamic demonstrations, the data fusion layer 3 completes the fusion of electrical, automatic control, analysis, and video data, and provides data support for the intelligent application layer 4. The intelligent application layer 4 includes functions such as automatic inspection 5, online equipment diagnosis and predictive maintenance 6, process flow simulation 7, online leakage monitoring 8, intelligent production management and control 9, data statistical analysis 10, and intelligent reporting 11. Automatic Inspection 5: Automatically acquires video along the route, sensor data, and pipeline leakage status through preset inspection paths to achieve automatic inspection in the virtual production scene; Online Equipment Diagnosis 6: Constructs fault diagnosis and prediction models for impeller wear, bearing failure, etc. by integrating temperature, vibration, and electrical parameters; Process Simulation 7: Data-driven simulation of processes such as hydrometallurgical ion exchange, material transport, and injection, real-time updates of equipment status, and visualization of real-time alarm information; Online Leakage Monitoring 8: Overlays online leakage monitoring data at corresponding locations on the 3D model of the wastewater treatment evaporation tank, accurately locating leakage situations and triggering alarms and emergency plans, achieving visualized online control of safety and environmental protection; Intelligent Production Management 9: A unified time-series library associated with 3D coordinates enables an interactive mode of clicking on the model to retrieve real-time data, and a 2D / 3D fusion interface enables visualized monitoring of the production process. Clicking on any device in the virtual scene allows for remote control; Data Statistical Analysis 10: Generates curves for material consumption, energy consumption statistics, etc. in real time; Intelligent Reporting 11: Allows flexible configuration of multi-dimensional data daily, monthly, and yearly reports for automatic control, analysis, and management, enabling multi-dimensional comparison, analysis, statistics, and output of production data.

[0047] This application provides a data monitoring system for uranium mine production processes, including a visualization subsystem, a data processing subsystem, and a data monitoring subsystem. The visualization subsystem constructs a semantic 3D model of the uranium mine production process based on uranium mine production components, and visualizes multi-source data based on this semantic 3D model. The semantic 3D model includes geometric elements, texture elements, semantic elements, and relational elements related to the production process. The data processing subsystem collects the multi-source data in real time, performs labeling and fusion on the multi-source data, and sends it to the data monitoring subsystem. The data monitoring subsystem performs fault diagnosis on the labeled and fused multi-source data based on a transfer learning algorithm, and displays the obtained fault diagnosis results in the visualization subsystem. This system achieves full intelligentization of the uranium mine production process, significantly improves data integration and monitoring effectiveness, reduces material and labor costs, and realizes integrated comprehensive management and control of online monitoring, remote equipment control, and data statistical analysis, thereby improving production safety and meeting the effective data monitoring needs of uranium mine production processes.

[0048] Obviously, those skilled in the art should understand that the modules or steps of this application described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented here, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, this application is not limited to any particular combination of hardware and software.

[0049] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A data monitoring system for uranium mine production processes, characterized in that, include: The system comprises a visualization subsystem, a data processing subsystem, and a data monitoring subsystem. The visualization subsystem is used to construct a semantic 3D model of the uranium mine production process based on uranium mine production components, and to visualize multi-source data based on the semantic 3D model. The semantic 3D model includes geometric elements, texture elements, semantic elements, and relational elements in the production process. The data processing subsystem is used to collect the multi-source data in real time, and to label and fuse the multi-source data before sending it to the data monitoring subsystem. The data monitoring subsystem is used to perform fault diagnosis on the multi-source data after label fusion based on the transfer learning algorithm, and to display the obtained fault diagnosis results in the visualization subsystem.

2. The system according to claim 1, characterized in that, The semantic 3D model includes a geometric sub-model and a process flow sub-model; The visualization subsystem is specifically used to add semantic information and rule function information to the uranium mine production components on the geometric sub-model, determine the process flow information of the uranium mine production process, and adjust the process flow sub-model according to the process flow information. The visualization subsystem is further configured to combine the geometric sub-model with the process flow sub-model, and mark geometric elements, texture elements, semantic elements and relational elements according to the semantic information, rule function information and process flow information to generate the semantic 3D model.

3. The system according to claim 2, characterized in that, The data processing subsystem is specifically used to acquire multi-source data collected by different multi-source devices through an integrated interface. The multi-source data includes at least one of image data, signal data, numerical data, and text data. The data processing subsystem is further used to extract features from the multi-source data, label the extracted data features, and store them in the database.

4. The system according to claim 3, characterized in that, The data processing subsystem is also used to retrieve data features that match the tags carried in the data retrieval instruction from the database when it receives a data retrieval instruction from the data monitoring subsystem. The data processing subsystem is also used to perform feature fusion on the data features and feed them back to the data monitoring subsystem.

5. The system according to claim 4, characterized in that, The data monitoring subsystem is also used to respond to detection commands, determine target markers, and send data retrieval commands. The data monitoring subsystem is also used to identify the data features of the target labeled as an image based on a visual recognition model to obtain visual diagnostic results; The data monitoring subsystem is also used to identify the data features of the target labeled as numerical or text based on the environmental identification model, and obtain environmental diagnosis results.

6. The system according to claim 5, characterized in that, The data monitoring subsystem is also used to associate and bind the three-dimensional coordinates of the semantic three-dimensional model with the data time series library, and to send an early warning command to the visualization subsystem when data points in the data time series library are diagnosed as abnormal. The visualization subsystem is also used to respond to the warning command by generating a warning area using the three-dimensional coordinates and visually displaying the warning area.

7. The system according to claim 6, characterized in that, The visualization subsystem is also used to obtain the multi-source data from the data processing subsystem, generate a production process consumption curve corresponding to the multi-source data, and perform time-series visualization early warning on the production process consumption curve through a consumption threshold.

8. The system according to claim 7, characterized in that, The data monitoring subsystem is also used to determine the inspection object based on distance and online status and generate an inspection path when the fault diagnosis result indicates that a fault has occurred. Send the inspection path to the inspection object to perform offline inspection.

9. The system according to claim 8, characterized in that, The data monitoring subsystem is also used to generate emergency decision-making strategies based on the fault diagnosis results and send them to the visualization subsystem to output the emergency decision-making strategies.

10. The system according to claim 9, characterized in that, The data monitoring subsystem is also used to acquire the multi-source data, generate statistical results of the multi-source data according to a preset time interval, and output them.

Images