A method of automatically calibrating the position of a gas valve of a sorter

Abstract

This invention relates to a method for automatically calibrating the position of an air valve in a pneumatic sorting machine system. The pneumatic sorting machine system includes a conveyor belt, an identification module, a spraying module, a sorting module, and an image analysis module. A high-frame-rate industrial camera is installed above and to the side of the spraying area. The method includes the following steps: S1, system startup; S2, detecting the ore position and calculating the air valve position; S3, starting the air valve spraying and the camera acquiring images of the spraying process; S4, time-series analysis of the spraying results; S5, calculation of the spraying deviation; S6, generation of spraying position calibration parameters; S7, parameter update; repeating steps S2 to S7 to form a closed-loop control, continuously bringing the spraying position parameters closer to the optimal spraying position, thus achieving automatic calibration of the air valve spraying position. This invention improves sorting accuracy and operational stability by real-time monitoring of the spraying process and automatically calibrating the spraying position based on the feedback of the spraying results.

Description

Technical Field

[0001] This invention belongs to the field of material sorting technology, and in particular relates to a method for automatically calibrating the position of the air valve of a sorting machine. Background Technology

[0002] Pneumatic separators are widely used in the automated sorting of materials such as ores, scrap metals, and recycled resources. These devices typically use sensors or vision recognition systems to detect and identify materials on the conveyor belt. Combining this information with the material's position and speed on the conveyor belt, they calculate the corresponding air valve number and the trigger time for the air jet when the target material reaches the jetting area. The jetting system then separates the target material from the main material stream. In existing separators, the jetting position usually includes the left-right correspondence of the air valves and the jetting timing parameters. Calibration is often done manually or through static calibration based on fixed operating conditions, and the equipment is put into use after initial installation or maintenance.

[0003] However, in actual operation, the aforementioned existing technical solutions still have significant shortcomings. On the one hand, under high-speed operation of the sorting machine, the movement posture, trajectory, and landing point of the material are easily affected by factors such as differences in material shape, speed fluctuations, and conveyor belt vibration, resulting in deviations between the actual blowing position and the pre-calibrated position. Once the air valve triggers prematurely, delayedly, or laterally, even if the identification result is accurate, problems such as blowing failure, misalignment, or incorrect blowing may occur, thereby reducing the sorting accuracy. On the other hand, during long-term operation, factors such as changes in the installation position of the air valve array, shifts in the image acquisition perspective, conveyor belt deviation, mechanical vibration, and structural micro-deformation may all cause the original blowing position calibration to gradually become invalid. Existing equipment still mainly relies on manual shutdown for readjustment, making it difficult to achieve real-time compensation during dynamic operation. This not only affects production continuity but also increases maintenance costs and reliance on operational experience. With the development of high frame rate industrial cameras, real-time image processing technology and high-speed actuators, sorting equipment has the capability to monitor the blowing effect in real time. However, existing technologies still lack an effective method to automatically calibrate the blowing position based on the difference between the expected blowing result and the actual blowing effect of the target object. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method for automatically calibrating the position of the air valve in a sorting machine. By monitoring the blowing process in real time and automatically calibrating the blowing position based on the feedback of the blowing results, the sorting accuracy and operational stability are improved.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: A method for automatically calibrating the position of the air valve of a sorting machine is applied to a pneumatic sorting machine system. The pneumatic sorting machine system includes a conveyor belt, an identification module, a blowing module, a sorting module, and an image analysis module. A high frame rate industrial camera is set above and to the side of the blowing area to form a field of view covering the blowing area, and image sequences of the blowing area and the material movement trajectory area are continuously acquired at a preset frame rate. The method includes the following steps: S1, system startup; S2, Detecting ore position and calculating valve position: The image of the recognition area is processed by the recognition module to obtain the outline, center position, movement speed and lateral coordinates of the target object; and based on the conveyor belt speed, material trajectory model and the currently calibrated injection position parameters, the target valve number and the trigger time of the target valve when the target object arrives at the injection area are predicted. S3, start the air valve to blow and the camera to collect images of the blowing process: control the blowing module to drive the corresponding air valve to blow according to the target air valve number and trigger time obtained in step S2, and at the same time, the high frame rate industrial camera continuously records the images of the entire blowing process to ensure that the blowing action and the image acquisition process are strictly synchronized. S4, Time-series analysis of the injection result: The image analysis module performs time-series analysis on the image of the injection area to determine whether the injection is accurate. Specifically, it determines whether there is a left or right lateral deviation in the actual injection position of the air valve based on the offset direction of the material in the injection area, and determines whether the air valve is injecting too early or too late based on the movement trajectory of the material. S5, Calculation of spray deviation: When premature spraying, delayed spraying, or lateral offset is detected, the spray deviation is calculated based on the image analysis results. S6, Generating injection position calibration parameters: The system generates injection position calibration parameters based on the injection deviation. The injection position calibration parameters include timing calibration factor, lateral calibration factor, and valve trigger mapping table update parameters. S7, Parameter Update: Automatically apply the updated jetting position calibration parameters to the next jetting process; Repeat steps S2 to S7 to form a closed-loop control, so that the injection position parameters continuously approach the optimal injection position, thereby achieving automatic calibration of the air valve injection position.

[0006] Preferably, the deviation in step S5 includes: valve trigger advance amount. Valve trigger delay Left / right lateral deviation The difference in valve numbers corresponding to the actual injection deviation .

[0007] Preferably, the valve number difference corresponding to the actual blowing deviation is determined based on the spatial distribution relationship between the left / right lateral deviation and the valve array in the blowing module.

[0008] Preferably, in step S4, if it is detected that there is no premature blowing, delayed blowing, or lateral offset in the blowing, the calibration is completed.

[0009] Preferably, in step S7, the update of the blowing position calibration parameter needs to meet at least one of the following: the update is performed only when the blowing deviation exceeds a preset threshold; the update is performed after statistically analyzing the blowing deviation based on multiple consecutive blowing results; or the update is performed after aggregating the blowing deviations of multiple targets using a sliding window.

[0010] Preferably, in step S7, when premature or delayed injection is detected, only the timing calibration factor is updated; when lateral deviation is detected, only the lateral calibration factor or the valve trigger mapping table update parameter is updated.

[0011] Preferably, in step S7, the update range of the blowing position calibration parameter each time is limited by a preset maximum adjustment amount to avoid over-correction of the blowing position.

[0012] Preferably, the injection position calibration parameter is used to correct the prediction model parameters for the target object reaching the injection zone, so that the subsequent prediction of the target object's valve number and trigger time is performed based on the corrected prediction model.

[0013] Preferably, the update frequency or update magnitude of the blowing position calibration parameters is dynamically adjusted according to the size, speed, or trajectory characteristics of the target object.

[0014] Preferably, the time-series analysis of the blowing results is cross-validated based on image information from high frame rate industrial cameras above and to the side to improve the accuracy of blowing deviation judgment.

[0015] Compared with the prior art, the advantages of the present invention are as follows: (1) This invention achieves automatic calibration of the air valve blowing position by real-time image acquisition of the blowing process in the blowing area, and calculates the blowing deviation and generates the blowing position calibration parameters based on the blowing results. It does not require manual shutdown for debugging or reliance on manual experience, and can effectively avoid blowing deviation caused by equipment installation errors, conveyor belt deviation, mechanical vibration or structural changes caused by long-term operation, thereby improving the automation level and operational reliability of the sorting system.

[0016] (2) This invention performs time-series analysis on the images of the blowing area, identifies the advance or delay deviation in the blowing sequence and the blowing offset of the air valve in the lateral direction, and constructs a two-dimensional calibration model based on time and space deviation. It then updates the timing calibration factor, lateral calibration factor and air valve trigger mapping relationship in a targeted manner, so that the blowing action can always be consistent with the actual position of the target object, thereby effectively reducing the phenomena of missed spraying, mis-spraying and spray deviation, and significantly improving the sorting accuracy and sorting stability.

[0017] (3) The present invention uses the blowing result itself as the calibration basis and constructs a closed-loop automatic calibration mechanism based on visual feedback, so that the blowing position calibration parameters can be continuously optimized according to the material size, movement speed, movement trajectory and working condition changes during equipment operation, realizing adaptive and self-learning dynamic calibration. This not only enhances the sorting machine's adaptability to complex working conditions and multiple types of materials, but also reduces the frequency of manual intervention, improves the continuous operation capability of the equipment, and reduces maintenance costs and usage threshold. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the pneumatic sorting machine system of the present invention; Figure 2 This is a top view schematic diagram of the pneumatic sorting machine system of the present invention; Figure 3 This is a flowchart of the method for automatically calibrating the position of the air valve of the sorting machine according to the present invention; Reference numerals: 1-Conveyor belt; 2-Identification module; 21-Detector; 22-X-ray source; 3-Puffing module; 4-Sorting module; 5-High frame rate industrial camera. Detailed Implementation

[0019] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention belong to the present invention.

[0020] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this application. However, those skilled in the art will recognize that the technical solutions of this application can be practiced without one or more of the specific details, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this application.

[0021] like Figure 1-3As shown, this embodiment discloses a method for automatically calibrating the position of the air valve of a sorting machine, which is applied to a pneumatic sorting machine system. The pneumatic sorting machine system includes a conveyor belt 1, an identification module 2, a blowing module 3, a sorting module 4, and an image analysis module. A high frame rate industrial camera 5 is set above and to the side of the blowing area to form a field of view covering the blowing area, and continuously acquires image sequences of the blowing area and the material movement trajectory area at a preset frame rate.

[0022] The conveyor belt 1 is used to carry and transport materials to be sorted. The identification module 2 is used to identify and detect the target objects on the conveyor belt 1. The identification module 2 includes a detector 21 and a radiation source 22. The detector 21 and the radiation source 22 are arranged in pairs above and below or on both sides of the conveyor belt 1 to detect the materials passing through the conveyor belt 1, such as raw ore, waste rock or other materials to be sorted. The radiation source 22 emits radiation signals to the materials. After the radiation signals pass through the materials, they are received by the corresponding detector 21. The identification module 2 identifies and analyzes the contour features, internal characteristics or position status of the materials based on the signal changes received by the detector 21, thereby providing basic data support for the positioning of the target objects, the calculation of their motion state and the valve triggering decision of the subsequent blowing module 3.

[0023] The blowing module 3 includes multiple air valves arranged along the conveying direction for blowing and sorting the target material. The sorting module 4 is used to collect the material after blowing. A high-frame-rate industrial camera 5 captures images of the blowing area from above and the side to ensure that the blowing action and material movement are completely recorded. The image analysis module analyzes and processes the acquired blowing process images, providing a data basis for blowing position calibration. In this embodiment, the image analysis module can be integrated into the controller of the pneumatic sorting machine system, using a processor or industrial control computing unit to perform real-time analysis and processing of the image data acquired by the high-frame-rate industrial camera 5.

[0024] The method includes the following steps: S1, System Start-up; In this step, the pneumatic sorting machine system enters the running state, and each module completes initialization and begins to work together. S2, Detecting ore position and calculating valve position: The image of the identification area is processed by the identification module 2 to obtain the outline, center position, movement speed and lateral coordinates of the target object. The identification module 2 can image or detect the target object based on the detector 21 and the X-ray source 22, and predict the target valve number and the trigger time of the target valve when the target object arrives at the injection area according to the conveyor belt speed, the pre-established material trajectory model and the currently calibrated injection position parameters, thereby providing a basis for subsequent injection control. S3, start the air valve to spray and the camera to collect images of the spraying process: control the spraying module 3 to drive the corresponding air valve to spray according to the target air valve number and trigger time obtained in step S2, so that the target object deviates from its original trajectory under the action of the spraying airflow. At the same time, the high frame rate industrial camera 5 continuously records the images of the entire spraying process to ensure strict synchronization between the spraying action and the image acquisition process in time, so as to accurately reflect the spraying effect. S4, Time-series analysis of the injection result: The image analysis module performs time-series analysis on the image of the injection area to determine whether the injection is accurate. Specifically, it determines whether there is a left or right lateral deviation in the actual injection position of the air valve based on the offset direction of the material in the injection area, and determines whether the air valve is injecting too early or too late based on the movement trajectory of the material. S5. Calculation of spray deviation: When premature spraying, delayed spraying, or lateral offset is detected, the spray deviation is calculated based on the image analysis results. This spray deviation is used to quantify the difference between the spraying action and the theoretical spraying position. S6. Generation of spray position calibration parameters: The system generates spray position calibration parameters based on the spray deviation. The spray position calibration parameters include a timing calibration factor for correcting the spray timing, a lateral calibration factor for correcting the lateral position of the spray, and a valve trigger mapping table update parameter for adjusting the correspondence between the air valve and the target object. S7. Parameter Update: The updated jetting position calibration parameters are automatically applied to the next jetting process, so that subsequent jetting control is executed based on the corrected parameters. Repeat steps S2 to S7 to form a closed-loop control, so that the injection position parameters continuously approach the optimal injection position during continuous injection, thereby achieving automatic calibration of the air valve injection position.

[0025] Furthermore, the deviation in step S5 includes: valve trigger advance. Valve trigger delay Left / right lateral deviation The difference in valve numbers corresponding to the actual injection deviation Among them, the valve trigger advance amount Gas valve trigger delay This is used to characterize the deviation of the jetting action relative to the target object's arrival at the jetting zone over time; the left / right lateral deviation. Used to characterize the degree of deviation of the jetting airflow relative to the center position of the target object in the lateral direction of space; the difference in valve number corresponding to the actual jetting deviation. This is used to reflect the mismatch in the selection of the injection valve due to lateral deviation; the valve number corresponding to the actual injection deviation. The difference is based on the left / right lateral deviation. The spatial distribution relationship with the air valve array in the injection module 3 is determined. By mapping the lateral offset distance with the spacing between adjacent air valves, the air valve number or air valve trigger mapping relationship that needs to be corrected is determined.

[0026] In this embodiment, in step S4, if it is detected that there is no premature blowing, delayed blowing, or lateral offset in the blowing, the calibration is completed. This indicates that the current blowing position parameters can keep the blowing action consistent with the target object position in time and space. The system does not need to update the blowing position calibration parameters and can maintain the existing parameters to continue to execute the subsequent blowing and sorting process.

[0027] In step S7, updating the injection position calibration parameters requires at least one of the following conditions: updating is only performed when the injection deviation exceeds a preset threshold; updating is performed after statistically analyzing the injection deviation based on multiple consecutive injection results; or updating is performed after aggregating the injection deviations of multiple targets using a sliding window. Setting a preset threshold avoids frequent adjustments due to occasional disturbances or single abnormal injections. Statistical analysis of multiple consecutive injection results improves the stability and reliability of calibration parameter updates. The sliding window selects the most recent injection results of several targets in the time series and aggregates the corresponding injection deviations, thereby suppressing the impact of random noise on the calibration process while ensuring response speed. Furthermore, in step S7, when injection is detected to be premature or delayed, only the timing calibration factor is updated to correct the injection trigger time to the actual arrival time of the target in the injection zone. When lateral deviation is detected, only the lateral calibration factor or the valve trigger mapping table update parameters are updated to realign the injection airflow with the center position of the target in lateral space, thus avoiding unnecessary parameter disturbances caused by the coupling of different types of deviations. Furthermore, in step S7, the update range of the calibration parameters for each injection position is limited by a preset maximum adjustment amount to avoid over-correction of the injection position in a single or a few calibration processes, ensuring a smooth and gradual calibration process, thereby improving the stability of injection control and the overall reliability of system operation.

[0028] In this embodiment, the injection position calibration parameters are used to correct the prediction model parameters for the target object's arrival at the injection zone. This allows subsequent predictions of the target object's valve number and trigger time to be performed based on the corrected prediction model. Specifically, by feeding back the time and spatial deviations obtained during the injection process to the target object's motion prediction model, the speed parameters of conveyor belt 1, the material trajectory model parameters, or the valve trigger mapping relationship are corrected. This allows subsequent targets to complete more accurate valve number selection and trigger time calculation based on the updated model before entering the injection zone. The update frequency or magnitude of the injection position calibration parameters is dynamically adjusted according to the target object's size, speed, or trajectory characteristics. For larger targets, faster targets, or targets with significant trajectory fluctuations, the update frequency or magnitude is increased to enhance the system's responsiveness to dynamically changing conditions. For targets with stable operating conditions, the update frequency or magnitude is decreased to ensure the overall stability of the system. The time-series analysis of the blowing results is based on cross-validation of image information from the high frame rate industrial camera 5 above and to the side. By jointly analyzing images of the same blowing process from different perspectives, misjudgments caused by single-view occlusion, imaging distortion or illumination changes are eliminated, thereby improving the accuracy and reliability of blowing deviation judgment.

[0029] This invention addresses the problems in existing sorting machines, such as fixed valve blowing positions, drift in blowing position calibration over time, and the tendency for premature blowing, delayed blowing, or incorrect valve blowing during actual sorting. It provides an automatic valve blowing position calibration method based on visual feedback. By utilizing a high-frame-rate industrial camera (5) to monitor the valve blowing process in real time, and comparing the actual blowing results with the expected blowing position of the target object, the method identifies the temporal and spatial deviations during blowing. This allows for automatic calibration and dynamic compensation of the blowing position, enabling continuous optimization of blowing control during equipment operation. This improves sorting accuracy and system stability, reduces the frequency of manual adjustments, and enhances the adaptability of the sorting equipment to different working conditions and material changes.

[0030] In summary, this invention discloses a method for automatically calibrating the position of air valves in a pneumatic sorting machine. By introducing a high-frame-rate industrial camera 5 into the pneumatic sorting machine system to collect real-time data on the blowing process, and combining this with an identification module 2 to predict and analyze the position and motion state of the target object on the conveyor belt 1, the method controls the blowing module 3 to perform the blowing action. Based on an image analysis module, the method performs temporal and spatial deviation analysis on the blowing results, calculates the blowing deviation, and generates blowing position calibration parameters. This allows for dynamic correction of subsequent blowing processes, forming a complete closed-loop automatic calibration mechanism. This technical solution can automatically identify blowing advance, delay, and lateral deviations during equipment operation, and achieve fine-tuning of the blowing position through temporal calibration factors, lateral calibration factors, and air valve trigger mapping table parameter updates. This significantly improves sorting accuracy and operational stability, reduces missed and incorrect blowing, and lowers the frequency of manual adjustments and maintenance costs. By using the injection results themselves as the calibration basis, this invention breaks through the technical limitations of existing sorting equipment that relies on static calibration and manual experience. It provides an intelligent sorting control method with adaptive and continuous optimization capabilities for fields such as ore sorting and renewable resource sorting. It has important engineering application value and industry promotion significance for promoting the development of sorting equipment towards automation and intelligence.

[0031] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Under the concept of the present invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the present invention as described above. For the sake of brevity, they are not provided in detail. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automatically calibrating the position of a pneumatic valve in a sorting machine, applied to a pneumatic sorting machine system, characterized in that, The pneumatic sorting machine system includes a conveyor belt (1), an identification module (2), a spraying module (3), a sorting module (4), and an image analysis module. A high frame rate industrial camera (5) is set above and to the side of the spraying area to form a field of view covering the spraying area and continuously collect image sequences of the spraying area and the material movement trajectory area at a preset frame rate. The method includes the following steps: S1, system startup; S2, Detect the position of the ore and calculate the position of the air valve: The image of the identification area is processed by the identification module (2) to obtain the outline, center position, movement speed and lateral coordinate of the target object; and according to the conveyor belt speed, material trajectory model and the currently calibrated injection position parameters, the target air valve number and the trigger time of the target air valve when the target object arrives at the injection area are predicted. S3, start the air valve to spray and the camera to collect images of the spraying process: control the spraying module (3) to drive the corresponding air valve to spray according to the target air valve number and trigger time obtained in step S2, while the high frame rate industrial camera (5) continuously records the images of the entire spraying process to ensure that the spraying action and the image acquisition process are strictly synchronized. S4, Time-series analysis of the injection result: The image analysis module performs time-series analysis on the image of the injection area to determine whether the injection is accurate. Specifically, it determines whether there is a left or right lateral deviation in the actual injection position of the air valve based on the offset direction of the material in the injection area, and determines whether the air valve is injecting too early or too late based on the movement trajectory of the material. S5, Calculation of spray deviation: When premature spraying, delayed spraying, or lateral offset is detected, the spray deviation is calculated based on the image analysis results. S6, Generating injection position calibration parameters: The system generates injection position calibration parameters based on the injection deviation. The injection position calibration parameters include timing calibration factor, lateral calibration factor, and valve trigger mapping table update parameters. S7, Parameter Update: Automatically apply the updated jetting position calibration parameters to the next jetting process; Repeat steps S2 to S7 to form a closed-loop control, so that the injection position parameters continuously approach the optimal injection position, thereby achieving automatic calibration of the air valve injection position.

2. The method for automatically calibrating the position of the air valve in a sorting machine according to claim 1, characterized in that, The deviation in step S5 includes: valve trigger advance amount. Valve trigger delay Left / right lateral deviation The difference in valve numbers corresponding to the actual injection deviation .

3. The method for automatically calibrating the position of the air valve in a sorting machine according to claim 2, characterized in that, The difference in valve number corresponding to the actual spraying deviation The determination is based on the relationship between the left / right lateral deviation and the spatial distribution of the air valve array in the injection module.

4. The method for automatically calibrating the position of the air valve in a sorting machine according to claim 3, characterized in that, In step S4, if it is detected that there is no premature blowing, delayed blowing, or lateral offset in the blowing process, the calibration is completed.

5. The method for automatically calibrating the position of the air valve in a sorting machine according to claim 1, characterized in that, In step S7, the update of the spray position calibration parameter needs to meet at least one of the following: the update is performed only when the spray deviation exceeds a preset threshold; the update is performed after statistically analyzing the spray deviation based on multiple consecutive spray results; or the update is performed after aggregating the spray deviations of multiple targets using a sliding window.

6. The method for automatically calibrating the position of the air valve in a sorting machine according to claim 5, characterized in that, In step S7, when premature or delayed injection is detected, only the timing calibration factor is updated; when lateral deviation is detected, only the lateral calibration factor or the valve trigger mapping table update parameter is updated.

7. The method for automatically calibrating the position of the air valve in a sorting machine according to claim 6, characterized in that, In step S7, the update range of the spray position calibration parameters is limited by a preset maximum adjustment amount to avoid over-correction of the spray position.

8. The method for automatically calibrating the position of the air valve in a sorting machine according to claim 1, characterized in that, The injection position calibration parameters are used to correct the prediction model parameters for the target object reaching the injection zone, so that the subsequent prediction of the target object's valve number and trigger time are based on the corrected prediction model.

9. The method for automatically calibrating the position of the air valve in a sorting machine according to claim 1, characterized in that, Based on the size, speed, or trajectory characteristics of the target object, the update frequency or magnitude of the spray position calibration parameters is dynamically adjusted.

10. The method for automatically calibrating the position of the air valve in a sorting machine according to claim 1, characterized in that, The time-series analysis of the blowing results is based on cross-validation of image information from high frame rate industrial cameras above and to the side to improve the accuracy of blowing deviation judgment.

Images