A clean coal flotation reagent regulation method and reagent regulation system thereof

Abstract

The application discloses a kind of clean coal flotation reagent regulation methods and its reagent regulation system, in clean coal flotation process, first by visual module record image sample file;Again by central processing module to sample file processing, obtains flotation state judgment reference set;Then the size of multiple foam objects in flotation state judgment reference set is calculated, and foam size distribution proportion chart of sample unit is drawn generation;Then according to the uniformity change trend of foam layer, the thickness change trend of foam layer in multiple continuous sample units, the relative proportion condition of the foam object in ideal foam size range in each sample unit and foam stability, the flotation effect in sampling time period is comprehensively judged, and the reagent adjustment addition amount is calculated and obtained;By adjusting dosing module, the reagent addition amount is real-time regulated according to the obtained reagent adjustment addition amount.The application can regulate reagent addition amount according to the real-time dynamic change of foam layer, to improve flotation effect and stability.

Description

Technical Field

[0001] This invention relates to the field of coal preparation technology, and in particular to a method for controlling flotation reagents for clean coal and a reagent control system thereof. Background Technology

[0002] Flotation washing of coal is a common coal preparation process. Various reagents need to be added during the process to obtain a better flotation effect. The amount of reagents added is usually determined by the analysis of the raw coal washability test. However, due to the complexity of the slurry composition, it is common for the amount of reagents added to be too much or too little in actual addition. Too much or too little reagent will seriously affect the flotation effect and cause instability in the flotation effect. Regular inspection and observation are usually required, and the amount of reagents added should be adjusted according to experience. Summary of the Invention

[0003] Purpose of the invention: In order to overcome the shortcomings of the existing technology, the present invention provides a method and system for regulating reagents in the flotation of clean coal, which can automatically regulate the amount of reagents added according to the real-time dynamic changes of the froth layer, so as to improve the flotation effect and stability.

[0004] Technical solution: To achieve the above objectives, the present invention provides a method for controlling the reagents used in clean coal flotation and a reagent control system thereof, which controls the actual amount of reagents to be added in real time during the clean coal flotation process.

[0005] The reagent control system includes: a vision module for real-time observation of the dynamic changes of the foam layer and acquisition of image samples; a central processing module for processing the information acquired from the image samples, analyzing the information to determine the flotation state, making corresponding decisions, and sending control commands; and an adjustment dosing module for executing control commands and controlling the real-time dosing amount.

[0006] The specific drug administration steps are as follows:

[0007] S1. The vision module records the dynamic changes of the foam layer and saves a historical image every preset time interval, using the latest saved image as the sample file.

[0008] S2. The central processing module processes the sample file, taking a single frame image as a sample unit, and taking multiple sample units with a preset frame interval within the sample file. Each sample unit records the foam layer thickness, foam morphology, and distribution state at the corresponding time. Image analysis is performed on each sample unit.

[0009] Multiple sub-samples are taken within a single sample unit, and feature extraction is performed on each of the multiple sub-samples. The multiple sub-samples are evenly distributed within the sample unit, and the distribution positions of the sub-samples within the multiple sample units overlap.

[0010] In a single subsample, extract the foam boundary contour, and use the area circled by the complete contour as a single foam object.

[0011] First, remove incomplete objects that are close to the border line in each sub-sample. Then, by comparing the size changes of the same foam object in two adjacent sample units, remove foam objects in each sample unit that have a large size change relative to the same foam object in the adjacent sample units on both sides. The set of remaining foam objects in each sample unit is used as the reference set for judging the flotation state of that sample unit.

[0012] S3. In a single sample unit, the size of multiple foam objects in the flotation state judgment reference set is calculated, the distribution pattern of the size of the multiple foam objects is summarized, and the degree of overlap with the relative ideal foam size range is determined, and a foam size distribution ratio diagram of the sample unit is drawn.

[0013] S4. First, determine the uniformity of the foam layer in the corresponding sample unit based on the slope of the distribution ratio of each foam size. Then, based on the trend of the uniformity of the foam layer, the trend of the thickness of the foam layer, the proportion of foam objects within the ideal foam size range relative to all foam objects in each sample unit, and the foam stability in multiple consecutive sample units, comprehensively judge the flotation effect within the sampling period. If the flotation effect is low, analyze the corresponding influencing factors and calculate the corresponding reagent adjustment and addition amount.

[0014] S5. The reagent addition amount of the clean coal flotation system is adjusted in real time by the reagent adjustment module according to the reagent adjustment amount obtained in step S4.

[0015] Further, in step S2, after removing objects with incomplete borders, the centroid of the remaining foam objects in the subsample is calculated first. Then, according to the direction of foam layer movement, the displacement distance of the same foam object in adjacent sample units is extracted to match and lock all the same foam objects in the corresponding subsamples in adjacent sample units. By comparing the locked foam objects, foam objects with an overlap degree lower than the preset value are removed.

[0016] Furthermore, in step S3, when calculating the size of a single foam object, the convex segments in the multi-segment contour lines of the foam object are first extracted, the center point of the convex segment is connected to the centroid of the foam object, the length of the multi-segment connection is calculated, and the average length of the multi-segment connection is taken as the radius of the foam object.

[0017] Furthermore, if the outline of the foam object does not include convex segments, then the center points of multiple straight outlines are connected to the centroid of the foam object, the lengths of the multiple connecting lines are calculated, and the average value of the lengths of the multiple connecting lines is taken as the radius of the foam object.

[0018] Furthermore, in step S1, the vision module takes a picture perpendicular to the plane where the foam layer is located, and the picture area is the foam scraping area far away from the flotation area of ​​the flotation tank.

[0019] Further, in step S4, the trend of foam layer thickness change is obtained by capturing and recording the marker array on both sides of the foam scraping area of ​​the flotation tank using a vision module. The marker array contains multiple marker units arranged equidistantly in the vertical direction. The number of marker units in each sample unit represents the foam layer thickness at the corresponding time. The trend of foam layer thickness change within the sample time period is obtained based on the number of markers in multiple consecutive sample units.

[0020] Furthermore, in step S4, the foam stability is obtained by the sound recognition module classifying the sound when the foam layer is scraped out according to the pre-stored sound reference samples, and then judging the foam stability according to the obtained sound type.

[0021] Furthermore, it includes an adjustment dosing module, which is equipped with multiple dosing pipelines corresponding to multiple flotation cells. One end of each dosing pipeline is connected to the impeller cavity of the aeration mechanism corresponding to the flotation cell, and the other end is connected to dosing tanks of various reagents through branch lines. Each branch line is equipped with a dosing regulating valve for controlling the flow rate.

[0022] Furthermore, each of the flotation cells is equipped with a vision module and a sound recognition module at the foam scraping outlet. The vision module and the sound recognition module are connected to multiple dosing regulating valves corresponding to the flotation cell via control signal transmission through a central processing module.

[0023] Furthermore, each flotation cell is equipped with a reagent concentration detection module at its slurry discharge outlet, and the dosing pipeline is equipped with a total quantity control valve. The reagent concentration detection module is connected to the total quantity control valve of the corresponding flotation cell via a control signal transmission from the central processing module.

[0024] Beneficial Effects: This invention provides a method and system for controlling reagents in the flotation of clean coal. Through a visual module, it observes the dynamic changes in the froth layer of each flotation cell in real time, generating corresponding reagent ratio control commands to fine-tune the reagent ratio in each flotation cell, maintaining the flotation effect of each cell. By detecting the reagent concentration at the pulp outlet of each flotation cell, it controls the total amount of reagent added to the next flotation cell, maintaining the reagent addition amount around or within a preset range. This allows for dynamic fine-tuning of each flotation cell without deviating from the preset total reagent addition amount. The reagent control is more precise, reliable, and timely, improving the flotation effect and stability. Attached Figure Description

[0025] Figure 1This is a block diagram of a method for controlling flotation reagents for clean coal according to the present invention;

[0026] Figure 2 This is a schematic diagram of the structure of a drug regulation system according to an embodiment of the present invention. Detailed Implementation

[0027] The invention will now be further described with reference to the accompanying drawings.

[0028] As attached Figure 1-2 The aforementioned method and system for controlling reagents in clean coal flotation allows for real-time control of the actual amount of reagents to be added during the clean coal flotation process.

[0029] The drug control system includes:

[0030] The vision module is used to observe the dynamic changes of the foam layer in real time and to acquire image samples.

[0031] The central processing module is used to process image sample acquisition information, analyze the information to determine the flotation state, make corresponding decisions, and send control commands.

[0032] The dosing module is used to execute control commands and regulate the real-time dosing amount.

[0033] The specific drug administration steps are as follows:

[0034] S1. The visual module records the dynamic changes of the foam layer, saving a historical image at preset intervals, with the most recently saved image serving as the sample file. The visual module takes images perpendicular to the plane of the foam layer, focusing on the foam scraping area far from the flotation zone of the flotation tank. Because bubbles continuously rise from the bottom of the flotation tank and randomly intersect within the foam layer, the dynamic changes of the foam layer are diverse and irregular. In contrast, the foam scraping area, relatively far from the flotation zone, does not receive direct bubble replenishment from the bottom; instead, bubbles are scraped from the surface of the flotation zone to the surface of the foam scraping area by a scraper. Most of the foam only undergoes lateral movement along the liquid surface, facilitating image analysis after sample collection and avoiding the increased image processing difficulty caused by bubbles emerging from uncertain locations.

[0035] The sample file is a monitoring video segment of a set duration. The specific duration can be determined through experimental research, starting from the time required for excessive or insufficient dosage of the agent to cause a significant change in the foam layer, and then setting a reasonable duration based on this. Furthermore, the sample file is time-sensitive; only the most recently saved file is used as the analysis sample, with 5-10 historical sample files cached for retrieval and viewing, and any extra files promptly removed. Real-time monitoring footage can also be retrieved, allowing monitors to directly observe the foam layer status. If an anomaly is detected during monitoring, manual intervention can be performed to adjust the dosage. After manual intervention, the timer is reset, and the sample file is saved for subsequent autonomous adjustments. In short, the system primarily relies on automatic control, but allows for random manual intervention during inspections, significantly reducing the frequency of manual inspections and the time spent on human observation.

[0036] S2. The central processing module processes the sample file, taking a single frame image as a sample unit. Multiple sample units with a preset frame interval are taken from the sample file. Each sample unit records the foam layer thickness, foam morphology, and distribution state at the corresponding time. Image analysis is performed on each sample unit. Each sample unit is a screenshot in the video sample file. Multiple screenshots are taken at fixed time intervals as multiple sample units.

[0037] Multiple sub-samples are taken within a single sample unit, and feature extraction is performed on each sub-sample. The sub-samples are evenly distributed within the sample unit, and their distribution positions overlap. The sub-samples are located in the foam layer area within the sample unit. A small area in the top view of the foam layer is defined by a rectangle of fixed length and width. Multiple sub-samples define multiple areas of the same size. These areas are equidistantly distributed along the direction perpendicular to the foam movement. The sub-samples within multiple sample units are selected at the same position, which can be used to compare the dynamic changes of the foam layer in this area.

[0038] In a single subsample, extract the foam boundary contour, and use the area circled by the complete contour as a single foam object.

[0039] First, remove incomplete objects that are close to the border line in each sub-sample. Then, by comparing the size changes of the same foam object in two adjacent sample units, remove foam objects in each sample unit that have a large size change relative to the same foam object in the adjacent sample units on both sides. The set of remaining foam objects in each sample unit is used as the reference set for judging the flotation state of that sample unit.

[0040] After removing objects with incomplete borders, the centroid of the remaining foam objects in the subsample is calculated. A matrix algorithm can be used to calculate the centroid based on the outline of each foam object.

[0041] Next, according to the direction of foam layer movement, the displacement distance of the same foam object in adjacent sample units is extracted. This is used to match and lock all identical foam objects within corresponding sub-samples in adjacent sample units. Foam objects locked by overlap comparison are then eliminated if the overlap is lower than a preset value. Specifically, the centroids of two foam objects in adjacent sample units whose contours are almost completely overlapped in the direction of foam layer movement are grouped together. Multiple groups are extracted in this way, and the distance between the centroids of each group is obtained. The average of several distance values ​​with relatively close values ​​is taken as the movement distance of the foam layer between the corresponding time points of the two sample units. The two foam objects whose centroids are grouped together are considered to be mutually matched.

[0042] Next, the remaining unmatched foam objects in the two sample units are matched. Specifically, based on the overlap of the images of two adjacent sample units, a semi-circular reference line is first drawn with the centroid of the unmatched foam object in the sample unit relative to the previous moment as the center and the calculated foam layer movement distance as the radius, moving in the direction of foam layer movement. The unmatched foam object in the sample unit relative to the next moment whose centroid position is closest to this reference line is considered the matching object. If the match is successful, the two foam objects are extracted according to their contours, and their centroids are made to coincide. The overlap of the two foam objects is then calculated. If the overlap meets the preset range, the two foam objects are placed into the floating state judgment reference set of the corresponding sample unit. If the overlap is too low, the two foam objects are removed. If the match is unsuccessful, that is, no matching centroid position is found around the semi-circular reference line, the foam object is directly removed. Based on the above matching process, in the final flotation state judgment reference set obtained by two adjacent sample units, each foam object can find a one-to-one correspondence with a high degree of overlap. This eliminates developing foam objects and individual unstable and broken foam objects, and uses foam objects with long existence time, stable structure and more complete structure as the judgment objects of foam layer state, making the judgment of foam layer state more reliable and representative.

[0043] S3. Within a single sample unit, the dimensions of multiple foam objects in the flotation state judgment reference set are calculated. The distribution patterns of the dimensions of these multiple foam objects and their overlap with the ideal foam size range are summarized, and a foam size distribution percentage map of the sample unit is generated. The foam size distribution percentage map uses percentage as the horizontal axis and foam size as the vertical axis, which can intuitively reflect the distribution of foam size in the foam layer.

[0044] When calculating the size of a single foam object, the convex segments of the multi-segment contour lines of the foam object are first extracted. The center point of each convex segment is then connected to the centroid of the foam object, and the length of this connecting line is calculated. The average length of this connecting line is taken as the radius of the foam object. Because the foam objects in the foam layer are relatively compressed, the foam has an irregular shape. The convex segments are in directions with relatively less compression and are closer to the original contour of the foam. Some convex segments may be further convex due to compression. Therefore, using the average length of the connecting lines from the center points of the multiple convex segments to the centroid as the radius value of the foam object is closer to the actual size of the foam object.

[0045] If the outline of the foam object does not contain a convex segment, then the center points of multiple straight outlines are connected to the centroid of the foam object, the length of the multiple connecting lines is calculated, and the average length of the multiple connecting lines is taken as the radius of the foam object.

[0046] If the outline of the foam object consists entirely of concave arcs, then the intersection of multiple concave arc segments is connected to the centroid, and the length of the connecting lines is calculated. The average length of the connecting lines is then taken as the radius of the foam object.

[0047] S4. First, determine the uniformity of the foam layer in the corresponding sample unit based on the slope of the distribution ratio of each foam size. Then, based on the trend of the uniformity of the foam layer, the trend of the thickness of the foam layer, the proportion of foam objects within the ideal foam size range relative to all foam objects in each sample unit, and the foam stability in multiple consecutive sample units, comprehensively judge the flotation effect within the sampling period. If the flotation effect is low, analyze the corresponding influencing factors and calculate the corresponding reagent adjustment and addition amount.

[0048] Specifically, when the total amount of foam objects in the flotation state judgment reference set decreases, it indicates that the amount of foam in the foam layer is reduced, and the foam is easily broken, resulting in fewer matching foam objects. This suggests that the content of collector in the reagent is too low, so the proportion of collector added is controlled to increase. If the above situation occurs at the same time, the proportion of large-sized foam objects in the flotation state judgment reference set increases, and the foam layer becomes thinner. This indicates that the amount of foam in the foam layer is low, it is easily broken, and it also contains large foam. This suggests that the amount of frother added is too low, so the proportion of frother added is controlled to increase.

[0049] When the total amount of foam objects in the flotation state judgment reference set increases and the proportion of small-sized foam increases significantly, it indicates that the content of collector is too high, so the proportion of collector added should be reduced. If the foam layer thickness increases significantly at the same time as the above situation, it indicates that the content of frother is too high, so the proportion of frother added should be reduced.

[0050] The ideal foam size range is generally 3-5cm, the large foam size range is generally 8-10cm, and the small foam size range is generally 1-2cm.

[0051] Based on the above identification of the state characteristics of the foam layer, there are some overlaps and differences in the characteristics of the foam layer changes caused by excessive or insufficient addition of collectors and frothers in the flotation reagents. The main differences lie in the changes in foam layer thickness and the presence of large bubbles. Since the state of the foam layer is quite complex, if significant differences cannot be identified, a sound recognition module can be used to distinguish whether the change in the content of the collector or frother is causing the foam layer state change. For example, when there is too much or too little frother, the foam is tough and elastic, or some bubbles are too large. In both cases, the foam is brittle and easily breaks, making it difficult to scrape off. Therefore, scraping the foam will produce a relatively loud cracking sound. When there is too much or too little collector, the foam is dense or has a low content and small size, resulting in a hissing cracking sound when the foam is scraped off. Under ideal flotation conditions, the foam will produce a slight hissing sound due to movement. Therefore, based on the above set of sound recognition observations of the foam layer characteristics, it is possible to verify and summarize which reagent needs to be adjusted, improving accuracy and reliability.

[0052] Based on this, and combined with the judgment of the uniformity of the foam layer, while adjusting the proportion of each reagent, the air intake and stirring speed are also adjusted in a coordinated manner to maintain an air intake and stirring speed that are appropriate for the amount of reagent added, thereby helping to control the thickness of the foam layer and maintain the stability of each flotation cell.

[0053] The thickness variation trend of the foam layer is obtained by capturing and recording the marker array on both sides of the foam scraping area of ​​the flotation tank using a vision module. The marker array contains multiple marker units arranged equidistantly in the vertical direction. The number of marker units in each sample unit represents the thickness of the foam layer at the corresponding time. The thickness variation trend of the foam layer within the sample time period is obtained based on the number of markers in multiple consecutive sample units.

[0054] The foam stability is obtained by a sound recognition module classifying the sound when the foam layer is scraped out based on pre-stored sound reference samples, and then judging the foam stability according to the obtained sound type.

[0055] S5. The dosing adjustment module adjusts the reagent addition amount of the clean coal flotation system in real time according to the reagent adjustment addition amount obtained in step S4. The dosing adjustment module is equipped with multiple dosing pipelines 2 corresponding to multiple flotation cells 1. One end of each dosing pipeline 2 is connected to the impeller cavity of the aeration mechanism 20 corresponding to the flotation cell 1, and the other end is connected to the dosing tanks 4 of various reagents through branch lines 3. Each branch line is equipped with a dosing regulating valve 5 for controlling the flow rate. The dosing regulating valves 5 of each branch line are controlled by instructions generated by the central processing module to control the flow rate of each reagent added to the dosing pipeline 2, thereby adjusting the dosing ratio of each reagent.

[0056] Each flotation cell 1 is equipped with a vision module 6 and a sound recognition module 7 at its foam scraper outlet 11. Both the vision module 6 and the sound recognition module 7 are connected to multiple dosing control valves 5 corresponding to the flotation cell 1 via a central processing module 8. This enables individual dynamic control of the reagent ratio in each flotation cell, ensuring that each flotation cell maintains good flotation performance.

[0057] In addition, each flotation cell 1 is equipped with a reagent concentration detection module 9 at its slurry outlet 12, and the dosing pipeline 2 is equipped with a total quantity regulating valve 10. The reagent concentration detection module 9 is connected to the total quantity regulating valve 10 of the corresponding flotation cell 1 via a control signal transmission from the central processing module 8. The total quantity regulating valve 10 is used to regulate the total amount of reagent added to each flotation cell. Before flotation, a batch of coal mine is usually subjected to a beneficiation test to determine the range of addition amounts of each flotation reagent. Based on this, while achieving dynamic fine-tuning of the reagent ratio in each flotation cell, it is also necessary to ensure that the total amount of reagent added should be limited to near or within the preset addition amount to guarantee the overall flotation effect. Therefore, by detecting the reagent concentration at the slurry outlet of each flotation cell, the total amount of reagent added in the next flotation cell is regulated, and the addition ratio of each reagent in the corresponding flotation cell is regulated according to the foam layer state of each flotation cell, thereby achieving reagent regulation of the system. Dynamic fine-tuning is performed on each flotation cell without deviating from the preset total amount of reagent added. Drug regulation is now more precise and reliable.

[0058] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the above principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method of regulating a fine coal flotation reagent, characterized by: In the clean coal flotation process, the actual amount of required reagent addition is real-time regulated by a reagent regulation system; The reagent regulation system comprises: a visual module for real-time observation of the dynamic changes of the froth layer and image sample collection; A central processing module for processing image sample information, analyzing information to determine the flotation state, and making corresponding decisions and sending control instructions; an adjustment dosing module for executing control instructions to control the adjustment of real-time dosing; The specific reagent regulation steps are as follows: S1, record the dynamic changes of the froth layer by the visual module, save a segment of historical images every preset time interval, and take the latest saved image segment as a sample file; S2, process the sample file by the central processing module, take a single image as a sample unit, take multiple sample units at a preset frame interval in the sample file, each sample unit records the froth layer thickness and froth shape and distribution state at the corresponding time, and each sample unit is subjected to image analysis; Take multiple sub-samples in a single sample unit and extract features from multiple sub-samples, multiple sub-samples are evenly distributed in the sample unit, and the distribution positions of sub-samples in multiple sample units coincide; In a single sub-sample, the froth boundary contour is extracted, and the area enclosed by the complete contour line is a single froth object; First, remove the incomplete objects in each sub-sample that are attached to the frame line, then compare the size changes of the same froth object in adjacent two sample units, remove the froth objects in each sample unit that have large size change amplitude relative to the same froth objects in the adjacent sample units on both sides, and take the set of remaining froth objects in each sample unit as the flotation state judgment reference set of the sample unit; S3, in a single sample unit, calculate the size of multiple froth objects in its flotation state judgment reference set, summarize the distribution rule of the size of multiple froth objects, and the coincidence degree relative to the ideal froth size range, and generate a froth size distribution proportion graph for the sample unit; S4, first, judge the uniformity of the froth layer in the corresponding sample unit according to the slope of each froth size distribution proportion graph; then, according to the uniformity change trend of the froth layer, the froth layer thickness change trend, the proportion of the froth objects in the ideal froth size range relative to all froth objects in each sample unit, and the froth stability, the flotation effect in the sampling time period is comprehensively judged, if the flotation effect is low, the corresponding influencing factors are analyzed, and the corresponding reagent adjustment amount is calculated; S5, the adjustment dosing module adjusts the reagent addition amount of the clean coal flotation system in real time according to the reagent adjustment amount obtained in step S4.

2. The method of claim 1, wherein: In step S2, after removing the incomplete objects in the frame, first, calculate the centroid of the remaining froth objects in the sub-sample, then extract the displacement distance of the same froth object in adjacent sample units according to the moving direction of the froth layer, and use it to match and lock all the same froth objects in the corresponding sub-sample in the adjacent sample units, and remove the froth objects with a coincidence degree lower than a preset value by overlapping comparison of the locked froth objects.

3. The method of claim 2, wherein the method is characterized by: In step S3, when calculating the size of a single foam object, the outer convex section in the multi-section contour line of the foam object is extracted first, the center point of the outer convex section and the center of mass of the foam object are connected, the length of the multi-section line is calculated, and the average of the lengths of the multi-section line is taken as the radius size of the foam object.

4. The method of claim 3, wherein: If the contour line of the foam object does not contain an outer convex section, the center points of the multi-section straight contour line and the center of mass of the foam object are connected respectively, the lengths of the multi-section lines are calculated, and the average of the lengths of the multi-section lines is taken as the radius size of the foam object.

5. The method of claim 4, wherein the method is characterized by: In step S1, the visual module is perpendicular to the plane where the foam layer is located, and the shooting area is the foam scraping area away from the air flotation area of the flotation tank.

6. The method of claim 5, wherein: In step S4, the acquisition of the thickness variation trend of the foam layer is achieved by recording the identification array on the two side tank walls of the foam scraping area of the flotation tank through the visual module, the identification array contains a plurality of identification units arranged equidistantly along the vertical direction, the number of identification units in each sample unit represents the thickness of the foam layer at the corresponding time, and the number of identifications in the continuous multiple sample units is used to acquire the thickness variation trend of the foam layer in the sample time period.

7. The method of claim 6, wherein the method is characterized by: In step S4, the acquisition of the stability of the foam is achieved by classifying the sound of the foam scraping through the sound recognition module according to the pre-stored sound reference sample, and then making corresponding foam stability judgment according to the obtained sound type.

8. The reagent control system for a method of controlling reagents in clean coal flotation according to any one of claims 1-7, characterized in that: The adjusting dosing module is provided with a plurality of dosing pipelines (2) corresponding to a plurality of flotation tanks (1), one end of the dosing pipeline (2) is connected to the impeller cavity of the aeration mechanism (20) corresponding to the flotation tank (1), the other end is connected to the dosing tank (4) of a plurality of reagents through a branch (3), and each branch is provided with a dosing adjusting valve (5) for controlling the flow.

9. The method of claim 8, wherein the method further comprises: determining a concentration of the coal floatation reagent in the coal floatation process; and adjusting the concentration of the coal floatation reagent in the coal floatation process based on the determined concentration of the coal floatation reagent. The visual module (6) and the sound recognition module (7) are provided at the foam scraping outlet (11) of each flotation tank (1), and the visual module (6) and the sound recognition module (7) are connected to the plurality of dosing adjusting valves (5) corresponding to the flotation tank (1) through the central processing module (8) for signal transmission.

10. The reagent control system for a method of controlling reagents in clean coal flotation according to claim 9, characterized in that: The flotation tank (1) is provided with a reagent concentration detection module (9), and the dosing pipeline (2) is provided with a total amount adjusting valve (10). The reagent concentration detection module (9) is connected to the total amount adjusting valve (10) corresponding to the flotation tank (1) through the central processing module (8) for signal transmission.

Images