Automatic medium adding method for medium warehouse of coal preparation plant, storage medium and electronic equipment

Abstract

The invention discloses an automatic medium adding method for a medium library of a coal preparation plant, a storage medium and electronic equipment, and the method comprises the steps: obtaining the medium library information of the medium library, and calculating the volume of a to-be-added concentrated medium, the mass of to-be-added magnetite powder and the volume of to-be-added clear water according to the medium library information; inputting the medium library information into a preset medium library digital twin model to obtain key information of the medium library; according to the coordinates of the highest point of the medium pile, a model prediction control algorithm is adopted to generate an optimal traveling track, and the traveling crane is controlled according to the real-time coordinates of the traveling crane, the traveling speed and the optimal traveling track; according to the mass of the magnetite powder to be added, the coordinates of the grab bucket, the opening and closing state of the grab bucket and the mass cumulant of the grabbed materials, direct medium discharging and / or differential medium discharging are / is executed; and if the medium placing in-place signal is received, a clean water pump and a blower device are controlled according to the volume of the clean water to be added. According to the invention, automatic medium adding of the coal preparation plant is realized, and control precision and efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of coal preparation plants, and in particular to an automatic medium adding method, storage medium and electronic equipment for a medium storage in a coal preparation plant. Background Art

[0002] The addition of medium to the coal preparation plant's medium reservoir is a key step in the heavy medium coal preparation process. The control of medium concentration directly affects the coal separation effect and product quality. Currently, the main technical problems in the control of medium addition to the coal preparation plant's medium reservoir are as follows:

[0003] (1) The existing media storage system mainly uses a single density sensor control method. The measurement of a single sensor is easily affected by external interference, resulting in insufficient measurement accuracy, delayed control response, and difficulty in responding to sudden changes in the production process in a timely manner.

[0004] (2) Traditional mediator control methods often rely on manual experience and regular sampling and testing. This method is not only labor-intensive, but also has a long manual sampling cycle, making it impossible to achieve real-time control. Moreover, manual judgment is subjective, making it difficult to ensure control accuracy.

[0005] (3) Existing technologies lack effective utilization of multi-source information. Various sensor data operate independently, failing to achieve information fusion. The control strategy is single and historical data cannot be fully utilized. The system lacks predictive capabilities and is difficult to respond to changes in working conditions in advance.

[0006] (4) The existing media library and media control system has a low degree of integration. The information between the subsystems is isolated and lacks coordination. The control decision-making lacks intelligent support and the equipment utilization rate is low. Summary of the Invention

[0007] The purpose of the present invention is to overcome the shortcomings of the existing coal preparation plant medium storage addition control that cannot realize automatic addition and has low accuracy and efficiency, and to provide a coal preparation plant medium storage automatic addition method, storage medium and electronic equipment.

[0008] The technical solution of the present invention provides a method for automatically adding medium to a medium storage in a coal preparation plant, comprising:

[0009] Step S1, obtaining media library information of the media library, and calculating the volume of concentrated medium to be added, the mass of magnetite powder to be added, and the volume of clean water to be added based on the media library information, wherein the media library information includes the remaining volume of the combined medium barrel, the initial density of the combined medium, the real-time sorting density of the heavy medium, and the optimal sorting density;

[0010] Step S2: Input the media library information into a preset digital twin model of the media library to obtain key information of the media library, including the coordinates of the highest point of the media pile, the coordinates of the flushing trough, the real-time coordinates of the vehicle, the driving speed, the coordinates of the grab bucket, the opening and closing status of the grab bucket, the cumulative mass of the grabbed materials, and the remaining amount of media in the flushing trough;

[0011] Step S3, generating an optimal driving trajectory using a model predictive control algorithm according to the coordinates of the highest point of the medium pile, and controlling the vehicle according to the real-time driving coordinates, the driving speed, and the optimal driving trajectory;

[0012] Step S4, performing direct placement and / or differential placement according to the mass of the magnetite powder to be added, the grab bucket coordinates, the opening and closing state of the grab bucket, and the cumulative mass of the grabbed material;

[0013] Step S5: If a signal indicating that the medium has been added is received, the clean water pump and the blower are controlled according to the volume of clean water to be added.

[0014] In one of the optional technical solutions, the calculation of the volume of concentrated medium to be added, the mass of magnetite powder to be added, and the volume of clean water to be added based on the media library information includes:

[0015] The volume of the concentrated medium to be added, the mass of the magnetite powder to be added, and the flow rate of the clean water to be added are calculated using the following formulas:

[0016]

[0017] Among them, V c is the volume of the concentrated medium to be added; V s is the volume of the combined medium in the preset combined medium barrel; V0 is the volume of the remaining medium in the combined medium barrel; ρ c is the density of the medium to be added, ρ t is the optimal sorting density, ρ0 is the initial density of the mixture; is the volume fraction of magnetite powder in the concentrated medium, ρ w is the water density, ρ m is the density of magnetite powder; M m V is the mass of the magnetite powder to be added; w This is the volume of clean water to be added.

[0018] In one of the optional technical solutions, the media library mathematical twin model is established using the following method:

[0019] Obtaining work area status information of the work area, including medium pile shape, medium pile height, concentrate tank level, concentrate barrel level, concentrate barrel level, clean water flow, added medium density, driving status, grab bucket status, personnel status, and grabbed material quality;

[0020] Establishing a three-dimensional rectangular coordinate system according to the three-dimensional spatial coordinates of the medium stack shape, the medium stack height, the concentrate tank liquid level, and the concentrate barrel material level;

[0021] Using a data fusion algorithm to fuse the concentrated medium tank level, the clean water flow rate, and the concentrated medium density to generate fusion information;

[0022] A deep learning algorithm is used to establish a digital twin model of the medium library for the three-dimensional rectangular coordinate system, the fusion information, the driving status, the grab status, the personnel status and the quality of the grabbed materials.

[0023] In one of the optional technical solutions, step S3 includes:

[0024] Step S301, generating the optimal driving trajectory including the target position coordinates using the model predictive control algorithm based on the coordinates of the highest point of the media stack and preset motion constraints, wherein the preset motion constraints include a prediction time domain, a control time domain, a running speed, and an acceleration;

[0025] Step S302 : Controlling the vehicle using a proportional-integral-differential algorithm based on the vehicle's real-time coordinates, the vehicle's speed, and the vehicle's optimal trajectory.

[0026] In one of the optional technical solutions, step S302 includes:

[0027] Step S3021, calculating the driving position difference between the real-time driving coordinates and the target position coordinates;

[0028] Step S3022, using an identification algorithm to calculate the proportional value, integral value, and differential value of the proportional-integral-differential algorithm based on the vehicle position difference;

[0029] Step S3023, generating the target position coordinates using a PID algorithm according to the proportional value, the integral value, and the differential value;

[0030] Step S3024: Control the vehicle according to the real-time coordinates, the vehicle speed, and the optimal vehicle trajectory.

[0031] In one of the optional technical solutions, step S4 includes:

[0032] When the absolute value of the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added is less than or equal to a preset error threshold, executing the direct placement;

[0033] When the accumulated mass of the grabbed material is greater than the mass of the magnetite powder to be added, and the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added is greater than the preset error threshold, the difference adjustment is performed according to the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added;

[0034] When the accumulated mass of the grabbed material is less than the mass of the magnetite powder to be added, and the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added is greater than the maximum addition amount of a single grab, the direct release is performed first, and then the difference release is performed.

[0035] In one of the optional technical solutions, the calculation formula for the grab opening and closing angle under the differential release condition is:

[0036]

[0037] Among them, θ max is the maximum opening and closing angle of the grab bucket; M target M is the mass of the magnetite powder to be added; current is the cumulative mass of the grabbed material.

[0038] In one of the optional technical solutions, step S5 further includes:

[0039] Step S6: When the remaining amount of the medium in the flushing tank is greater than a preset remaining amount threshold, a cyclic flushing is performed until the remaining amount of the medium in the flushing tank is lower than the preset remaining amount threshold.

[0040] The technical solution of the present invention also provides a computer-readable storage medium, which stores computer instructions. When a computer executes the computer instructions, it is used to execute all the steps of the automatic medium addition method for the medium library of a coal preparation plant as described above.

[0041] The technical solution of the present invention further provides an electronic device, comprising:

[0042] at least one processor; and,

[0043] a memory communicatively connected to the at least one processor; wherein,

[0044] The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to execute the automatic medium addition method for the medium reservoir of a coal preparation plant as described above.

[0045] The above technical solution has the following beneficial effects: by obtaining the medium library information of the medium library and calculating the volume of concentrated medium to be added, the mass of magnetite powder to be added, and the volume of clean water to be added based on the medium library information, the medium library information is input into the preset medium library digital twin model to obtain key information of the medium library, including the coordinates of the highest point of the medium pile, the coordinates of the flushing trough, the real-time coordinates of the driving vehicle, the driving speed, the coordinates of the grab bucket, the opening and closing status of the grab bucket, the cumulative mass of the grabbed material, and the remaining amount of medium in the flushing trough; according to the coordinates of the highest point of the medium pile, the model predictive control algorithm is used to generate the optimal driving trajectory, and the driving is controlled according to the real-time coordinates of the driving vehicle, the driving speed, and the optimal driving trajectory; according to the mass of the magnetite powder to be added, the coordinates of the grab bucket, the opening and closing status of the grab bucket, and the cumulative mass of the grabbed material, direct medium addition and / or differential medium addition are executed; if a medium addition signal is received, the clean water pump and the blower are controlled according to the volume of clean water to be added, thereby realizing automatic medium addition in the coal preparation plant medium library and improving control accuracy and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The disclosure of the present invention will become more easily understood with reference to the accompanying drawings. It should be understood that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of the present invention. In the drawings:

[0047] Figure 1 A flowchart of an automatic medium adding method for a medium reservoir in a coal preparation plant provided by one embodiment of the present invention;

[0048] Figure 2 The present invention provides a hardware structure diagram of an electronic device for automatically adding medium to a medium storage in a coal preparation plant according to an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0050] It is easy to understand that according to the technical solution of the present invention, a variety of structural modes and implementation modes can be replaced with each other by those skilled in the art without changing the essential spirit of the present invention. Therefore, the following specific embodiments and drawings are only exemplary descriptions of the technical solution of the present invention and should not be regarded as the entire invention or as a limitation or restriction of the technical solution of the invention.

[0051] The directional terms such as up, down, left, right, front, back, front, back, top, and bottom mentioned or possibly mentioned in this specification are defined relative to the structure shown in the drawings. They are relative concepts and may vary depending on the location and usage of the device. Therefore, these or other directional terms should not be interpreted as restrictive.

[0052] like Figure 1As shown, an embodiment of the present invention provides a method for automatically adding medium to a medium storage in a coal preparation plant, comprising:

[0053] Step S1: Obtaining the media library information of the media library, and calculating the volume of the concentrated medium to be added, the mass of the magnetite powder to be added, and the volume of the clean water to be added based on the media library information. The media library information includes the remaining volume of the combined medium barrel, the initial density of the combined medium, the real-time separation density of the heavy medium, and the optimal separation density;

[0054] Step S2: Input the media library information into the preset media library digital twin model to obtain key information of the media library, including the coordinates of the highest point of the media pile, the coordinates of the flushing trough, the real-time coordinates of the driving vehicle, the driving speed, the coordinates of the grab bucket, the opening and closing status of the grab bucket, the cumulative mass of the grabbed materials, and the remaining amount of media in the flushing trough;

[0055] Step S3: Generate an optimal driving trajectory using a model predictive control algorithm based on the coordinates of the highest point of the medium pile, and control the vehicle based on the real-time coordinates, driving speed, and the optimal driving trajectory;

[0056] Step S4: performing direct feeding and / or differential feeding according to the mass of the magnetite powder to be added, the grab bucket coordinates, the grab bucket opening and closing state, and the accumulated mass of the grabbed materials;

[0057] Step S5: If a signal indicating that the medium has been added is received, the clean water pump and the blower are controlled according to the volume of clean water to be added.

[0058] Specifically, the present invention can be applied to electronic devices with processing capabilities, such as industrial controllers, such as distributed control systems (DCS). Preferably, the present invention is applied to programmable logic controllers (PLCs).

[0059] First, when it is necessary to add medium to the media reservoir, the controller executes step S1 to obtain the media reservoir information of the media reservoir, and calculates the volume of concentrated medium to be added, the mass of magnetite powder to be added, and the volume of clean water to be added based on the media reservoir information. The media reservoir information includes the remaining medium volume in the combined medium barrel, the initial density of the combined medium, the real-time sorting density of the heavy medium, and the optimal sorting density.

[0060] Among them, the medium refers to qualified media.

[0061] Then, step S2 is executed to input the media library information into the preset media library digital twin model to obtain key information about the media library. The media library digital twin model can dynamically update the media library's three-dimensional spatial model and key information based on real-time multi-source data input, reflecting the actual operation of the media library in real time. The media library digital twin model can monitor the media library's operating status in real time. For example, it can display the shape changes of the media pile, the movement trajectories of the driving crane and grab bucket, the material grabbing and delivery status, and the real-time values ​​of various process parameters. This provides operators and control systems with intuitive and accurate monitoring information, allowing them to promptly identify and address potential problems.

[0062] The digital twin model of the media library utilizes learned data features and inherent patterns to predict the operating status of the media library. Based on current operating status and historical data, the digital twin model can predict the library's future development trends, such as the media stack height, vehicle movement paths, material addition quantities, and potential equipment failures. These early predictions provide a basis for production scheduling and equipment maintenance, enabling proactive measures to optimize production processes, improve production efficiency, and increase equipment reliability.

[0063] Then, step S3 is executed, and the Model Predictive Control (MPC) algorithm is used to generate the optimal driving trajectory based on the coordinates of the highest point of the medium pile. The MPC algorithm predicts the system behavior in the future based on the mathematical model and current state of the system, and generates the optimal driving trajectory by optimizing the control input, and controls the driving according to the real-time coordinates, driving speed and optimal driving trajectory.

[0064] Next, step S4 is executed to perform direct placement and / or differential placement according to the mass of the magnetite powder to be added, the grab bucket coordinates, the grab bucket opening and closing state, and the accumulated mass of the grabbed materials.

[0065] Direct addition means that when the mass of the magnetite powder to be added matches the cumulative mass of the grabbed materials, there is no need to add any more magnetite powder, and the grabbed magnetite powder is directly added to the concentrated medium tank. Differential addition means that only the difference between the mass of the magnetite powder to be added and the cumulative mass of the grabbed materials is added to the concentrated medium tank, without having to add all the magnetite powder grabbed in the last time.

[0066] Finally, step S5 is executed. After the direct and / or differential addition processes are complete, the controller receives a signal indicating that the water has been added. It then adjusts the motor speed of the clean water pump by varying the motor power supply frequency based on the volume of clean water to be added, achieving precise control of the clean water pump flow rate to meet the water addition requirements under different operating conditions. Furthermore, by gradually increasing the motor power supply voltage and frequency, the motor accelerates from a static state to its rated speed in a balanced manner, avoiding current and mechanical shock during startup and extending the service life of the equipment.

[0067] The frequency adjustment range is 20-50Hz, allowing the clean water pump motor speed to be adjusted within a wide range to meet different clean water flow requirements. For example, when a larger flow rate is required, the frequency can be adjusted to a higher value; when a smaller flow rate is required, the frequency can be adjusted to a lower value.

[0068] At the same time, the controller controls the speed of the blower by detecting the uniformity of the medium to stir the medium so that the various components in the medium are fully mixed to ensure the uniformity of the medium.

[0069] The uniformity of the medium is measured using a multi-point online density measurement method. Multiple density sensors are set at different positions of the medium tank to measure the medium density at each point in real time. By comparing the density values ​​of each measuring point, the average density deviation is calculated. When the average density deviation of each measuring point is less than the preset concentration deviation, it is determined that the mixing is uniform, and the speed of the blower is reduced or maintained; when the average density deviation is greater than or equal to the preset concentration deviation, the speed of the blower is increased and the stirring force is increased until the medium is uniform. The preset concentration deviation can be calibrated according to actual conditions. For example, the preset concentration deviation can be 0.1g / cm 3 .

[0070] The blowing device may be configured to start after the clean water pump is started for a preset period of time, and the blowing device may be configured to stop after the clean water pump is stopped for a preset period of time, such as 30 seconds.

[0071] In this embodiment, media reservoir information is obtained from the media reservoir, and the volume of concentrated medium to be added, the mass of magnetite powder to be added, and the volume of clean water to be added are calculated based on the media reservoir information. The media reservoir information is then input into a preset digital twin model of the media reservoir to obtain key information of the media reservoir, including the coordinates of the highest point of the media pile, the coordinates of the media flushing trough, the real-time coordinates of the vehicle, the driving speed, the coordinates of the grab bucket, the opening and closing status of the grab bucket, the cumulative mass of the grabbed material, and the remaining amount of media in the media flushing trough. An optimal driving trajectory is generated based on the coordinates of the highest point of the media pile using a model predictive control algorithm. The vehicle is controlled based on the real-time coordinates, driving speed, and the optimal driving trajectory. Direct media addition and / or differential media addition are executed based on the mass of the magnetite powder to be added, the coordinates of the grab bucket, the opening and closing status of the grab bucket, and the cumulative mass of the grabbed material. If a media addition signal is received, the clean water pump and the blower are controlled based on the volume of clean water to be added, thereby achieving automatic media addition in the coal preparation plant media reservoir and improving control accuracy and efficiency.

[0072] In one embodiment, the volume of concentrated medium to be added, the mass of magnetite powder to be added, and the volume of clean water to be added are calculated based on the media library information, including:

[0073] The following formulas are used to calculate the volume of concentrated medium to be added, the mass of magnetite powder to be added, and the flow rate of clean water to be added:

[0074]

[0075] Among them, V c is the volume of concentrated medium to be added; V s is the preset volume of the combined medium tank; V0 is the remaining volume of the medium in the combined medium tank; ρ c is the density of the medium to be added, ρ t is the optimal sorting density, ρ0 is the initial density of the medium; is the volume fraction of magnetite powder in the concentrated medium, ρ w is the water density, ρ m is the density of magnetite powder; M m is the mass of magnetite powder to be added; V w The volume of clean water to be added.

[0076] Specifically, according to the law of conservation of volume: V s =V c +V0, and conservation of mass: ρ0V0+ρ c V c =ρ t (V0+V c ), calculate the volume of concentrated medium to be added V c and the density of the medium to be added ρ c , and then according to the volume fraction of magnetite powder in the concentrated medium Water density ρ w and magnetite powder density ρm Calculate the mass M of magnetite powder to be added m and the volume of clean water to be added V w , further improving the accuracy.

[0077] In one embodiment, the media library mathematical twin model is established using the following method:

[0078] Obtain the working area status information of the working area, including the shape of the medium pile, the height of the medium pile, the liquid level of the concentrated medium tank, the material level of the concentrated medium barrel, the liquid level of the concentrated medium barrel, the clean water flow rate, the density of the added medium, the driving status, the grab status, the personnel status, and the quality of the grabbed material;

[0079] A three-dimensional rectangular coordinate system is established based on the three-dimensional spatial coordinates of the medium pile shape, medium pile height, concentrated medium tank level, and concentrated medium barrel level;

[0080] The data fusion algorithm is used to fuse the concentrated medium tank level, clean water flow and concentrated medium density to generate fusion information;

[0081] A deep learning algorithm is used to establish a digital twin model of the medium library based on the three-dimensional rectangular coordinate system, fusion information, driving status, grab status, personnel status and quality of the grabbed materials.

[0082] Specifically, a 360° panoramic LiDAR system is installed on the top of the media storage system, creating a LiDAR scanning system. This 360° panoramic LiDAR system provides a comprehensive, all-around scan of the media storage environment, acquiring the three-dimensional coordinates of objects such as the media stack, concentrate tank, and concentrate barrel. The LiDAR scanning system detects the shape and height of the media stack, the liquid level of the concentrate tank, and the material level of the concentrate barrel, generating real-time 3D point cloud data. This 3D point cloud data contains a large number of spatial coordinate points on the surfaces of various objects within the storage system, accurately reflecting their geometry and position. A 3D rectangular coordinate system is established based on this collected 3D point cloud data, providing a unified reference framework for subsequent data processing and analysis. Various methods can be used to establish this 3D rectangular coordinate system, such as selecting a fixed point within the storage system as the origin and using the length, width, and height of the storage system as the X, Y, and Z axes, respectively. Each point in the 3D point cloud data is then mapped to this coordinate system, facilitating subsequent calculations and analysis. By monitoring the shape and height of the media pile, the accumulation of the media pile can be understood in real time, providing a basis for media addition and management. Monitoring the liquid level in the concentrated media tank and the material level in the concentrated media barrel helps to accurately control the storage and use of the media, avoiding overflow or shortage of the media and ensuring the smooth progress of the coal preparation process.

[0083] Among them, the data processing of the lidar scanning system includes:

[0084] Filtering: Filter the collected 3D point cloud data to remove noise points and outliers, improving data purity and accuracy. For example, methods such as median filtering and Gaussian filtering can be used to remove isolated points and unreasonable data points caused by environmental interference or equipment errors.

[0085] Registration: 3D point cloud data collected at different times or from different perspectives is registered to form a complete 3D spatial data set in the same coordinate system. The registration process typically involves steps such as feature extraction, matching, and transformation. For example, algorithms such as the Iterative Closest Point (ICP) algorithm are used to align multiple point cloud datasets to obtain a unified 3D spatial point cloud model.

[0086] Reconstruction: Based on the registered 3D point cloud data, a 3D spatial model of the media reservoir is generated using a 3D reconstruction algorithm. Common reconstruction methods include voxelization and surface fitting, which can transform discrete point cloud data into a continuous 3D geometric model that intuitively reflects the shape, structure, and spatial distribution of the media reservoir.

[0087] At the same time, a high-definition industrial camera array is deployed within the work area to create a visual camera system. This system clearly captures detailed information within the work area, such as the motion of the crane, the working status of the grab bucket, the activities of on-site personnel, and the operational status of the equipment. The camera array can be strategically positioned based on the size and layout of the work area to ensure comprehensive monitoring of the entire work area. Real-time image data is collected, and image processing techniques are used for target detection and tracking, motion parameter calculation, and anomaly identification. The target detection algorithm identifies target objects such as the crane, grab bucket, and personnel in the image, annotating them with bounding boxes and class information. The tracking algorithm continuously tracks the target objects and extracts their motion parameters, such as real-time coordinates, velocity, and acceleration. Furthermore, analysis of the image sequence can identify anomalies such as equipment failures and improper human operation. Real-time monitoring of the crane and grab bucket motion status optimizes scheduling and control strategies, improving operational efficiency. Monitoring on-site personnel activities helps ensure safety and prevent accidents. Monitoring equipment operational status allows for the timely detection and resolution of equipment failures, ensuring continuous and stable production. Among them, the high-definition industrial camera array uses industrial cameras with a resolution of not less than 1920×1080 and a frame rate of not less than 30fps, and has infrared night vision function to support all-weather monitoring.

[0088] Among them, the visual system data processing includes:

[0089] Object detection: Object detection algorithms are applied to image data collected by the visual camera system to identify target objects such as vehicles, grab buckets, and on-site personnel. Object detection algorithms such as YOLO and SSD can quickly and accurately locate and identify each object in the image, annotating it with a bounding box and category information.

[0090] Tracking algorithms: Based on target detection, tracking algorithms are used to continuously track the target object and extract its motion parameters, such as real-time coordinates, velocity, and acceleration. Tracking algorithms such as Kalman filtering and particle filtering can predict the target's next position based on its position changes in consecutive image frames, thereby enabling real-time monitoring of the target's motion state.

[0091] An S-shaped tension sensor is installed at the connection between the grab bucket and the wire rope, creating a tension sensing and measurement system. This system monitors material weight in real time, calculates the weight of each grab, and accumulates the total media addition to control the media release rate and achieve precise quantitative control. Simultaneously, by controlling the grab bucket's media release action and speed, the amount of material released each time is precisely controlled, ensuring the accuracy and stability of the media addition process. Thus, by real-time monitoring of material weight and cumulative media addition, the amount of media added can be precisely controlled, avoiding over- or under-addition, improving the accuracy and quality of coal preparation. Controlling the media release rate and achieving precise quantitative control helps optimize the media addition process, improve production efficiency, and reduce production costs.

[0092] Among them, the S-type tension sensor has a measurement range of 0-5000kg and an accuracy level of 0.1.

[0093] Among them, the data processing of the tension sensor measurement system includes:

[0094] Filtering: Filtering is performed on the weighing data to remove signal interference caused by sensor noise, environmental vibration, and other factors, thereby improving data stability and reliability. Filtering methods can use digital filters, such as low-pass filters, to remove high-frequency noise components and retain valid weighing signals.

[0095] Calibration: Filtered weighing data is calibrated to correct measurement errors caused by sensor drift, equipment aging, and other factors, providing accurate material quality information. The calibration process typically requires the use of calibration tools such as standard weights to establish a data calibration model and correct and compensate the weighing data.

[0096] Multiple sensors are installed around the concentrate tank to create a distributed sensor network. These sensors include level sensors, flow sensors, and density sensors. Level sensors measure the liquid level in the concentrate tank in real time, preventing overflow or shortages and ensuring a stable supply of media. Flow sensors monitor the flow of clean water within the pipeline, helping to optimize the media delivery process, improve efficiency, and reduce energy consumption. Density sensors measure the density of the media to ensure that its quality meets requirements and enhance coal preparation effectiveness. By monitoring and controlling the concentrate tank liquid level, clean water flow, and added media density, refined management of the media addition process is achieved, improving production efficiency and product quality in the coal preparation plant.

[0097] Among them, the distributed sensor network data processing includes:

[0098] Data fusion algorithms: Data fusion algorithms are used to fuse and process multi-source data collected by the distributed sensor network, such as the concentrate tank level, the added medium clean water flow rate, and the added medium density, achieving information complementarity. Data fusion algorithms, such as Kalman filter fusion and Dempster-Shafer evidence theory fusion, comprehensively consider the characteristics and reliability of each sensor data, eliminate redundancy and inconsistencies between data, and obtain more accurate and comprehensive information, providing reliable data support for subsequent modeling and control.

[0099] Then, a digital twin model of the media depot is constructed using deep learning algorithms, combining the 3D rectangular coordinate system, fused information, crane status, grab bucket status, operator status, and the quality of grabbed materials. Deep learning algorithms, such as convolutional neural networks (CNN), recurrent neural networks (RNN), and long short-term memory (LSTM), can automatically learn the complex features and inherent patterns in the data, constructing a highly realistic digital twin model of the media depot. Once the digital twin model is established, the operating status of the depot is monitored in real time. Based on real-time multi-source data input, the digital twin model dynamically updates the depot's 3D spatial model and key information, reflecting the depot's actual operating conditions in real time. For example, it can display the shape changes of the media pile, the motion trajectories of the crane and grab bucket, the material grabbing and delivery status, and the real-time values ​​of various process parameters. This provides intuitive and accurate monitoring information to operators and the control system, enabling timely identification and resolution of potential problems. Furthermore, the digital twin model of the media library utilizes learned data features and inherent patterns to predict the operating status of the media library. Based on current operating status and historical data, the digital twin model can predict the library's future development trends, such as the media stack height, vehicle movement paths, material addition quantities, and potential equipment failures. These early predictions provide a basis for production scheduling and equipment maintenance, enabling proactive measures to optimize production processes, improve production efficiency, and increase equipment reliability.

[0100] The input to the digital twin model of the media depot includes multi-source data collected in real time, such as the three-dimensional spatial coordinate system generated by the lidar, equipment motion parameters extracted by the vision system, material quality information obtained by the tension sensor measurement system, and comprehensive information fused from the distributed sensor network. This input data covers various aspects of the media depot's geometry, equipment operating status, material quality, and process parameters, providing a comprehensive and rich data foundation for the model's establishment. The output of the digital twin model of the media depot is key information about the media depot, such as the coordinates of the highest point of the media pile, the coordinates of the flushing trough, the real-time coordinates and speed of the driving crane, the coordinates and opening and closing status of the grab bucket, the accumulated mass of the grabbed material, the remaining amount of media in the flushing trough, and the personnel safety zone. This key information is the core indicator of the media depot's operating status and can provide a direct basis for subsequent control decisions, enabling precise monitoring and control of the media depot.

[0101] In one embodiment, step S3 includes:

[0102] Step S301: Based on the coordinates of the highest point of the medium pile and preset motion constraints, a model predictive control algorithm is used to generate an optimal driving trajectory including the target position coordinates. The preset motion constraints include the prediction time domain, the control time domain, the operating speed, and the acceleration.

[0103] Step S302: Based on the real-time coordinates, driving speed and optimal driving trajectory, a proportional-integral-differential algorithm is used to control the driving.

[0104] Specifically, the controller uses the model predictive control (MPC) algorithm to plan the trajectory based on the coordinates of the highest point of the medium pile, and obtains the optimal driving trajectory by optimizing the control input.

[0105] Preferably, the present invention adopts a rolling optimization method, that is, in each control cycle, the optimal control sequence for a period of time in the future is recalculated based on the latest system status and prediction model, and only the first control action of the sequence is implemented. This method can adjust the control strategy in real time, adapt to changes in system status, and improve the flexibility and robustness of control.

[0106] The prediction horizon refers to the time range within which the model predictive control algorithm predicts the future behavior of the system. Within this range, the algorithm will calculate a series of possible control inputs and the corresponding system states; the control horizon refers to the time range within which the control actions are actually implemented. That is, in each control cycle, only the control actions for the first 10 seconds are implemented based on the prediction results.

[0107] During trajectory planning and motion control, the prediction time domain is preferably 30s, the control time domain is preferably 10s, the speed is preferably no more than 2m / s, and the acceleration is preferably no more than 0.5m / s. 2 , ensuring the safety and stability of crane operation and preventing equipment damage, material spillage, or accidents caused by excessive speed or acceleration. By introducing these constraints during the optimization process, the MPC algorithm can achieve optimal trajectory control while meeting safety requirements, allowing the crane to quickly and smoothly reach the designated location.

[0108] The MPC algorithm can achieve optimal trajectory control of the vehicle. In each control cycle, the optimal trajectory for the next 30 seconds is calculated based on the coordinates of the highest point of the medium pile and the current driving status, and the corresponding control actions are implemented within the first 10 seconds. At the same time, the constraints of speed and acceleration are met, so that the vehicle can move smoothly and quickly along the predetermined optimal trajectory to the target position, providing a good foundation for subsequent precise positioning.

[0109] In one embodiment, step S302 includes:

[0110] Step S3021: Calculating the driving position difference between the real-time driving coordinates and the target position coordinates;

[0111] Step S3022: Calculate the proportional value, integral value, and differential value of the proportional-integral-differential algorithm using an identification algorithm based on the vehicle position difference;

[0112] Step S3023: Generate the target position coordinates using the PID algorithm according to the proportional value, integral value and differential value;

[0113] Step S3024: Control the vehicle according to the real-time coordinates, driving speed and optimal driving trajectory.

[0114] Specifically, the controller has a control cycle of 50 milliseconds. Every 50 milliseconds, the controller calculates and outputs a control signal based on the deviation between the current and target positions, adjusting the vehicle's motion in real time. This shorter control cycle improves control response speed and accuracy, enabling the vehicle to more quickly adapt to changes in position deviation and achieve precise positioning control.

[0115] PID parameters are updated in real time through online identification. This involves continuously adjusting the PID parameters during the control process based on the system's real-time input and output data to identify the optimal PID parameters. These parameters include Kp, Ki, and Kd. For example, by analyzing information such as the vehicle's position deviation and speed changes, identification algorithms such as least squares and recursive least squares can be used to calculate the most appropriate Kp, Ki, and Kd values, providing a basis for real-time adjustment of the control parameters. Real-time adjustment of control parameters ensures that the PID algorithm consistently maintains optimal control performance under varying operating conditions. For example, as the vehicle approaches the target position, appropriately reducing the Kp value and increasing the integral effect can improve positioning accuracy. Conversely, as the vehicle moves away from the target position, increasing the Kp value and strengthening the proportional effect can increase vehicle speed and shorten positioning time. This adapts to changing system characteristics and improves control performance.

[0116] Among them, Kp is preferably 0.5-2.0, Ki is preferably 0.1-0.5, and Kd is preferably 0.05-0.2.

[0117] In one embodiment, step S4 includes:

[0118] When the absolute value of the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added is less than or equal to the preset error threshold, direct addition is executed;

[0119] When the accumulated mass of the grabbed material is greater than the mass of the magnetite powder to be added, and the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added is greater than the preset error threshold, the difference adjustment is performed according to the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added;

[0120] When the accumulated mass of the grabbed material is less than the mass of the magnetite powder to be added, and the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added is greater than the maximum addition amount of a single grab, direct release is performed first, and then the difference release is performed.

[0121] Specifically, when the absolute difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added is less than or equal to the preset error threshold, the mass of the grabbed material and the mass of the magnetite powder to be added are essentially consistent, and the error is within an acceptable range. No complex adjustments are required at this point; the grabbed material can simply be placed in the designated location. This simple and quick operation effectively improves media addition efficiency. For example, if the preset error threshold is 50kg, the accumulated mass of the grabbed material is 1000kg, and the mass of the magnetite powder to be added is 1050kg, the absolute difference is 50kg, which is equal to the preset error threshold. Therefore, direct media placement is executed, and the grab bucket opening angle is the maximum.

[0122] When the cumulative mass of the grabbed material is greater than the mass of the magnetite powder to be added and the difference is greater than the preset error threshold, it indicates that the mass of the grabbed material exceeds the mass of the magnetite powder to be added, and the excess is large. At this time, it is necessary to partially release the grabbed material to match the actual added material mass with the required mass to avoid excessive material addition and affect the subsequent coal preparation effect.

[0123] If the cumulative mass of the captured material is less than the mass of the magnetite powder to be added, and the difference between the cumulative mass of the captured material and the mass of the magnetite powder to be added is greater than the maximum addition amount for a single grab, this indicates that the captured material mass is insufficient to meet the mass of the magnetite powder to be added, and the gap is large. Repeated direct media release is required, with multiple capture and release operations until the cumulative mass of the captured material exceeds the mass of the magnetite powder to be added and the difference exceeds the preset error threshold. Then, the difference release is performed to ensure the accuracy of the media addition amount and guarantee coal preparation quality. For example, if the preset error threshold is 50kg, the cumulative mass of the captured material is 900kg, and the mass of the magnetite powder to be added is 1000kg, the difference is 100kg, which is greater than the preset error threshold. Repeated capture and release operations are performed until the cumulative addition amount reaches approximately 1000kg, and then the difference release is performed.

[0124] In this embodiment, by accurately comparing the cumulative mass of the grabbed material with the mass of the magnetite powder to be added, and selecting the appropriate medium addition action according to the difference and the preset error threshold, it is possible to achieve precise control of the material addition amount, ensure the accuracy and stability of medium addition, and help improve the accuracy and quality of coal preparation; and it is possible to flexibly select the medium addition action according to different situations, and can effectively deal with whether the material mass is slightly more, slightly less, or the difference is large, with strong adaptability and applicable to various complex production conditions; when the material quality meets the requirements, the medium is directly added, reducing unnecessary operating steps; when the material quality does not meet the requirements, precise adjustment is made through differential medium addition or cyclic medium addition, avoiding material waste and excessive processing, improving production efficiency, and reducing production costs.

[0125] In one embodiment, the calculation formula for the grab bucket opening and closing angle under the differential release condition is:

[0126]

[0127] Among them, θ max is the maximum opening and closing angle of the grab bucket; M target M is the mass of magnetite powder to be added; current It is the cumulative amount of grabbed materials.

[0128] In one embodiment, step S5 further includes:

[0129] When the remaining amount of the medium in the flushing tank is greater than a preset remaining amount threshold, a cyclic flushing is performed until the remaining amount of the medium in the flushing tank is lower than the preset remaining amount threshold.

[0130] Specifically, since magnetite powder easily adheres to the inner wall of the medium storage tank, circulating flushing is performed by real-time detection of the remaining amount of medium (magnetite powder) in the medium storage tank, thereby circulating flushing the adhered medium until the remaining amount is lower than the preset remaining amount threshold.

[0131] Among them, circulating medium flushing refers to transporting the pre-prepared concentrated medium to the top of the medium placing tank through a circulating pump, so that the pre-prepared concentrated medium is flushed down from the top of the medium placing tank, taking away the magnetite powder adhering to the inner wall of the medium placing tank, and repeating this cycle until the magnetite powder on the inner wall of the medium placing tank is lower than the preset remaining amount threshold.

[0132] In one embodiment, step S5 further includes:

[0133] When a person is detected entering a dangerous area, an alarm is automatically sounded and the equipment is stopped.

[0134] Specifically, a visual camera system, constructed using an array of high-definition industrial cameras positioned throughout the work area, monitors on-site personnel activities in real time. Using image recognition and analysis technology, it accurately identifies personnel locations, movements, and whether they have entered pre-defined danger zones. When the visual camera system detects a person entering a danger zone, a controller automatically issues an alarm. This alarm can be either audible or visual, such as a high-decibel siren and flashing lights, both in the control room and on-site, to quickly attract the attention of personnel and operators, prompting them to evacuate the danger zone. Simultaneously, the controller automatically stops the operation of related equipment. For example, if a person enters the danger zone near the crane track, the crane will immediately stop. If a person approaches the grab bucket's operating range, the grab and release operations will cease, and the associated hydraulic or motor drive systems will be locked to prevent accidental equipment operation and possible injury.

[0135] It should be noted that in the media warehouse, dangerous areas are pre-set according to the equipment operation characteristics and safety regulations. These areas are usually places where people should not approach when the equipment is running, such as the driving track, the grab operation range, and the area near the concentrated medium barrel addition port. Entering these areas may cause safety accidents.

[0136] An embodiment of the present invention provides a computer-readable storage medium, which is used to store computer instructions. When a computer executes the computer instructions, it is used to execute all steps of the automatic medium addition method in the coal preparation plant medium library in any of the above-mentioned method embodiments.

[0137] like Figure 2 As shown in FIG. 1 , a hardware structure diagram of an electronic device for automatically adding medium to a medium storage in a coal preparation plant according to an embodiment of the present invention includes:

[0138] at least one processor 201; and,

[0139] A memory 202 in communication with at least one processor 201; wherein,

[0140] The memory 202 stores instructions that can be executed by at least one processor 201. The instructions are executed by at least one processor 201 so that the at least one processor 201 can execute the automatic medium addition method for the medium reservoir of the coal preparation plant in any of the above method embodiments.

[0141] Figure 2 A processor 201 is taken as an example.

[0142] The electronic device is preferably a programmable logic controller (PLC).

[0143] The electronic device may further include an input device 203 and an output device 204 .

[0144] The processor 201, the memory 202, the input device 203 and the output device 204 may be connected via a bus or other means, with the bus connection being used as an example in the figure.

[0145] The memory 202 is a non-volatile computer-readable storage medium that can be used to obtain non-volatile software programs, non-volatile computer executable programs and modules, such as the program instructions / modules corresponding to the automatic media addition method of the coal preparation plant media library in the embodiment of the present application, for example, Figure 1 The processor 201 executes various functional applications and data processing by running the non-volatile software programs, instructions and modules acquired from the memory 202, that is, realizing the automatic medium addition method of the medium depot in the coal preparation plant in the above embodiment.

[0146] Memory 202 may include a program acquisition area and a data acquisition area. The program acquisition area may acquire the operating system and at least one application required for a function; the data acquisition area may acquire data created based on the use of the automatic media loading method for the coal preparation plant media library. Furthermore, memory 202 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state memory device. In some embodiments, memory 202 may optionally include memory remotely located relative to processor 201. Such remote memory may be connected to a device executing the automatic media loading method for the coal preparation plant media library via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0147] The input device 203 can receive user clicks and generate signal input related to user settings and function control of the automatic medium addition method of the coal preparation plant medium library. The output device 204 can include a display device such as a display screen.

[0148] The one or more modules are acquired in the memory 202 and, when run by the one or more processors 201 , execute the method for automatically adding media to the media library of a coal preparation plant in any of the above method embodiments.

[0149] The above-mentioned product can execute the method provided in the embodiment of this application, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method provided in the embodiment of this application.

[0150] The above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, rather than to limit them. Although the embodiments of the present invention have been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automatically adding medium to a medium storage in a coal preparation plant, characterized in that: include: Step S1, obtaining media library information of the media library, and calculating the volume of concentrated medium to be added, the mass of magnetite powder to be added, and the volume of clean water to be added based on the media library information, wherein the media library information includes the remaining volume of the combined medium barrel, the initial density of the combined medium, the real-time sorting density of the heavy medium, and the optimal sorting density; Step S2: Input the media library information into a preset digital twin model of the media library to obtain key information of the media library, including the coordinates of the highest point of the media pile, the coordinates of the flushing trough, the real-time coordinates of the vehicle, the driving speed, the coordinates of the grab bucket, the opening and closing status of the grab bucket, the cumulative mass of the grabbed materials, and the remaining amount of media in the flushing trough; Step S3, generating an optimal driving trajectory using a model predictive control algorithm according to the coordinates of the highest point of the medium pile, and controlling the vehicle according to the real-time driving coordinates, the driving speed, and the optimal driving trajectory; Step S4, performing direct placement and / or differential placement according to the mass of the magnetite powder to be added, the grab bucket coordinates, the opening and closing state of the grab bucket, and the cumulative mass of the grabbed material; Step S5: If a signal indicating that the medium has been added is received, the clean water pump and the blower are controlled according to the volume of clean water to be added.

2. The automatic medium adding method for the medium storage of a coal preparation plant according to claim 1, characterized in that: The method of calculating the volume of concentrated medium to be added, the mass of magnetite powder to be added, and the volume of clean water to be added according to the medium library information includes: The volume of the concentrated medium to be added, the mass of the magnetite powder to be added, and the flow rate of the clean water to be added are calculated using the following formulas: Among them, V c is the volume of the concentrated medium to be added; V s is the volume of the combined medium in the preset combined medium barrel; V0 is the volume of the remaining medium in the combined medium barrel; ρ c is the density of the medium to be added, ρ t is the optimal sorting density, ρ0 is the initial density of the mixture; is the volume fraction of magnetite powder in the concentrated medium, ρ w is the water density, ρ m is the density of magnetite powder; M m V is the mass of the magnetite powder to be added; w This is the volume of clean water to be added.

3. The automatic medium adding method for the medium storage of a coal preparation plant according to claim 1, characterized in that: The medium library mathematical twin model is established using the following method: Obtaining work area status information of the work area, including medium pile shape, medium pile height, concentrate tank level, concentrate barrel level, concentrate barrel level, clean water flow, added medium density, driving status, grab bucket status, personnel status, and grabbed material quality; Establishing a three-dimensional rectangular coordinate system according to the three-dimensional spatial coordinates of the medium stack shape, the medium stack height, the concentrate tank liquid level, and the concentrate barrel material level; Using a data fusion algorithm to fuse the concentrated medium tank level, the clean water flow rate, and the concentrated medium density to generate fusion information; A deep learning algorithm is used to establish a digital twin model of the medium library for the three-dimensional rectangular coordinate system, the fusion information, the driving status, the grab status, the personnel status and the quality of the grabbed materials.

4. The automatic medium adding method for the medium storage of a coal preparation plant according to claim 1, characterized in that: The step S3 comprises: Step S301, generating the optimal driving trajectory including the target position coordinates using the model predictive control algorithm based on the coordinates of the highest point of the media stack and preset motion constraints, wherein the preset motion constraints include a prediction time domain, a control time domain, a running speed, and an acceleration; Step S302 : Controlling the vehicle using a proportional-integral-differential algorithm based on the vehicle's real-time coordinates, the vehicle's speed, and the vehicle's optimal trajectory.

5. The automatic medium adding method for the medium storage in a coal preparation plant according to claim 4, characterized in that: The step S302 includes: Step S3021, calculating the driving position difference between the real-time driving coordinates and the target position coordinates; Step S3022, using an identification algorithm to calculate the proportional value, integral value, and differential value of the proportional-integral-differential algorithm based on the vehicle position difference; Step S3023, generating the target position coordinates using a PID algorithm according to the proportional value, the integral value, and the differential value; Step S3024: Control the vehicle according to the real-time coordinates, the vehicle speed, and the optimal vehicle trajectory.

6. The automatic medium adding method for the medium storage of a coal preparation plant according to claim 1, characterized in that: The step S4 comprises: When the absolute value of the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added is less than or equal to a preset error threshold, executing the direct placement; When the accumulated mass of the grabbed material is greater than the mass of the magnetite powder to be added, and the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added is greater than the preset error threshold, the difference adjustment is performed according to the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added; When the accumulated mass of the grabbed material is less than the mass of the magnetite powder to be added, and the difference between the accumulated mass of the grabbed material and the mass of the magnetite powder to be added is greater than the maximum addition amount of a single grab, the direct release is performed first, and then the difference release is performed.

7. The automatic medium adding method for the medium storage in a coal preparation plant according to claim 6, characterized in that: The calculation formula for the grab opening and closing angle under the differential release condition is: Among them, θ max is the maximum opening and closing angle of the grab bucket; M target M is the mass of the magnetite powder to be added; current is the cumulative mass of the grabbed material.

8. The automatic medium adding method for the medium storage of a coal preparation plant according to any one of claims 1 to 7, characterized in that: The step S5 further includes: Step S6: When the remaining amount of the medium in the flushing tank is greater than a preset remaining amount threshold, a cyclic flushing is performed until the remaining amount of the medium in the flushing tank is lower than the preset remaining amount threshold.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and when a computer executes the computer instructions, it is used to execute all steps of the method for automatically adding medium to the medium storage of a coal preparation plant as described in any one of claims 1-8.

10. An electronic device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the automatic medium addition method for the medium storage of a coal preparation plant as described in any one of claims 1-8.

Images