Online detection and metering method applied to coal washing and dressing system

Abstract

The invention discloses an on-line detection and metering method applied to a coal washing system, which comprises the following steps: conveying raw coal in a coal receiving machine to a classifying screen, carrying out particle size detection on oversize and undersize of the classifying screen, and when the particle size of the undersize is not uniform, sending an alarm message that the classifying screen is abnormal; crushing detection is conducted on oversize products, crushed large raw coal blocks are selected to be subjected to secondary crushing, and when the raw coal needing secondary crushing exceeds a preset range, alarm information indicating that a crusher is abnormal is sent; conveying the raw coal subjected to secondary crushing and the screen underflow to a coal washer for classification to obtain gangue and middling coal; the middling coal is subjected to secondary screening through the jigger to obtain clean coal with smaller particle size, when the particle size of the clean coal subjected to secondary screening does not meet the preset requirement, the coal subjected to secondary screening is put into the jigger to be subjected to third-time screening, and when the third-time screening does not meet the preset requirement, alarm information about jigging abnormity is sent; and the clean coal meeting the preset requirement is metered.

Description

Technical Field

[0001] The present invention relates to the technical field of coal washing detection and metering, and in particular to an online detection and metering method applied to a coal washing system. Background Art

[0002] Washing coal is a process of finely treating raw coal using physical and other methods to improve the quality of clean coal products and reduce the ash, sulfur and volatile matter in clean coal to meet the customer's requirements for clean coal quality. This method of coal processing can separate and reduce undesirable components in coal, such as ash and sulfur, improve the energy utilization rate and economic value of coal, and reduce pollutants generated during combustion. Washing coal has become an important technology in the modern coal industry. It can not only improve the performance of fuel and reduce environmental pollution, but also promote the diversification of energy combinations and ensure energy security.

[0003] However, in the process of coal washing, how to cut the coal to achieve a balanced size of coal blocks fed into the jig, avoid damaging the jig, and extend the working life of the jig has become a major problem; in addition, different coal qualities have different requirements for washing processes. When dealing with certain complex coal qualities, existing technologies may find it difficult to achieve ideal coal preparation efficiency and clean coal quality. Efficient washing equipment has a high initial investment, and daily operation and maintenance also require high costs. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides an online detection and metering method applied to a coal washing system, the method comprising:

[0005] The raw coal in the coal receiving machine is transferred to the grading screen, and the particle size of the material on the grading screen and the material under the grading screen is detected. When the particle size of the material under the grading screen is uneven, an alarm message indicating the abnormality of the grading screen is sent;

[0006] The crushing detection is carried out on the screen material, and the large pieces of raw coal after crushing are selected for secondary crushing. When the raw coal that needs secondary crushing exceeds the preset range, an alarm message of crusher abnormality is sent;

[0007] The secondary crushed raw coal and the undersize are sent to the coal washer for classification to obtain gangue and medium coal;

[0008] The medium coal is screened twice by the jig to obtain clean coal with a smaller particle size. When the particle size of the clean coal screened twice does not meet the preset requirements, the coal after the secondary screening is put into the jig for a third screening. When the third screening still does not meet the preset requirements, an alarm message of abnormal jig is sent;

[0009] Measure the clean coal that meets the preset requirements.

[0010] Optionally, the raw coal in the coal receiving machine is transferred to a grading screen, and the particle size of the material on the grading screen and the material under the grading screen is detected. When the particle size of the material under the grading screen is uneven, an alarm message indicating that the grading screen is abnormal is sent. The specific contents include:

[0011] Extract the contour features of the raw coal and generate a cutting plan;

[0012] The cut raw coal is conveyed to the grading screen through the coal receiving machine;

[0013] The coal under the grading screen is tested for coal mine particle size using lifting wavelet and LBF model. When the test does not meet the particle size requirements, an abnormal alarm for the grading screen is sent.

[0014] Optionally, the method for automatically generating a cutting plan specifically includes:

[0015] Use superpixel algorithm to segment coal mine images and generate superpixel blocks;

[0016] Merging the super pixel blocks using a merging strategy, and scoring the coal block image to be processed according to the region cutting;

[0017] Labeling data of some of the coal block images to be processed, and constructing a segmentation model based on semi-supervised learning self-training according to the labeled data and the remaining data of the coal block images to be processed;

[0018] The segmentation model based on semi-supervised learning and self-training is used to cut the coal mine image to achieve the cutting of the coal mine.

[0019] Optionally, the coal under the grading screen is tested for coal mine particle size using lifting wavelet and LBF model. When the test fails to meet the particle size requirements, the content of the grading screen abnormality alarm is specifically sent including:

[0020] The cubic B-spline function is used as the prediction operator of lifting wavelet to filter and remove noise and detect edges of the acquired coal block image to be processed;

[0021] Perform morphological operations on the wavelet transformed image to eliminate the holes in the coal mine image.

[0022] The LBF model is used to cut the image after wavelet processing, and the connected city operation is performed on the cut image to obtain the number of coal mines. The number of pixels in the image is converted into the actual ore area, and the horizontal projection of the ore is equivalent to a circle. The approximate ore particle size is obtained through the area of ​​the circle.

[0023] The average value of the approximate ore particle size is obtained. If the average value of the approximate ore particle size exceeds the preset ore particle size threshold range, an abnormal alarm of the grading screen is issued.

[0024] Optionally, the crushing detection is performed on the material on the screen, and large pieces of raw coal are selected for secondary crushing after crushing. When the raw coal that needs secondary crushing exceeds a preset range, the alarm information of the crusher abnormality is sent, and the content specifically includes:

[0025] The U-NET network is used to extract the contour features of the large pieces of material on the sieve to obtain the first contour image;

[0026] Performing edge enhancement processing on the first contour image using a Canny algorithm to obtain a second contour image;

[0027] Calculate the large pieces of raw coal whose area blocks are larger than a preset threshold in the second contour map and put them into the crusher for secondary crushing;

[0028] If the area of ​​the large piece of raw coal in the second contour image is larger than the preset threshold range, an alarm message indicating the crusher is abnormal is sent.

[0029] Optionally, a U-NET network is used to extract contour features of the bulk oversize material, and the content of the first contour image obtained specifically includes:

[0030] When using U-NET, add an attention mechanism to the jump connection of the U-Net model to extract features;

[0031] The loss function is used to optimize the feature extraction content and obtain the first contour image.

[0032] Optionally, the secondarily crushed raw coal and the undersize are conveyed to a coal washer for classification to obtain gangue and medium coal, specifically including:

[0033] The gangue includes pure gangue and first medium coal, and its density is greater than 1.6g / cm 3 ;

[0034] The medium coal includes the second medium coal and clean coal, and its density is less than 1.6g / cm 3 .

[0035] Optionally, the medium coal is subjected to secondary screening by a jig to obtain clean coal with a smaller particle size. When the particle size of the clean coal after the secondary screening does not meet the preset requirements, the coal after the secondary screening is put into the jig for a third screening. When the third screening still does not meet the preset requirements, the content of the alarm information of the jig abnormality is sent, which specifically includes:

[0036] The size of the clean coal is tested. If the size meets the preset requirements, it is judged that there is no abnormality. If the size of the clean coal exceeds the preset requirements, a secondary screening is performed. If the size of the clean coal after the secondary screening still does not meet the preset requirements, the tool is judged to be abnormal, the jig operation is stopped, and a fault signal is issued.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] The present invention uses a two-time cutting method. First, a semi-supervised method is used to plan and cut the raw coal. Subsequently, contour cutting is performed based on the raw coal image after transportation and cutting to obtain a clearer raw coal image, so that the size of the cut coal blocks is more uniform. The present invention monitors the working status of each component in real time. When a problem occurs in a component and causes a problem in the coal washing system, it can be discovered and maintained in time. Finally, the metering method of the present invention can predict the future output rate of clean coal and monitor the operating status of the entire coal washing system based on the output. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0040] Figure 1 This is a method step diagram of an online detection and metering method applied to a coal washing system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] The raw coal in the coal receiving machine is transferred to the grading screen, and the particle size of the sieve material and the sieve material of the grading screen is detected. When the particle size of the sieve material is uneven, an alarm message of abnormal grading screen is sent; Optionally, the raw coal in the coal receiving machine is transferred to the grading screen, and the particle size of the sieve material and the sieve material of the grading screen is detected. When the particle size of the sieve material is uneven, an alarm message of abnormal grading screen is sent. The specific contents include: extracting the contour features of the raw coal and generating a cutting plan; transferring the cut raw coal to the grading screen through the coal receiving machine; using the lifting wavelet and LBF model to detect the coal mine particle size of the coal under the grading screen, and sending an abnormal grading screen alarm when the detection does not meet the particle size requirements. In the present invention, the particle size requirement range is 8-13mm.

[0044] The method for automatically generating a cutting plan specifically includes: using a superpixel algorithm to segment a coal mine image to generate superpixel blocks; using a merging strategy to merge the superpixel blocks, and scoring the coal block image to be processed according to the regional cutting score; marking the data of part of the coal block image to be processed, and constructing a segmentation model based on semi-supervised learning self-training according to the marked data and the data of the remaining coal block images to be processed; using the segmentation model based on semi-supervised learning self-training to cut the coal mine image to achieve the cutting of raw coal.

[0045] The coal under the grading screen is tested for coal mine particle size using lifting wavelet and LBF model. When the test does not meet the particle size requirements, the content of sending the grading screen abnormal alarm specifically includes: using cubic B-spline function as the prediction operator of lifting wavelet to filter and denoise the acquired coal block image to be processed and detect the edge; performing morphological operation on the image after wavelet transformation to eliminate the hole part of the coal block image to be processed; performing image cutting using LBF model on the image after wavelet processing, performing connected city operation on the cut image to obtain the number of coal mines, converting the number of pixels in the image into the actual ore area, and equating the horizontal projection of the ore to a circle, and obtaining the approximate ore particle size through the area of ​​the circle; obtaining the mean of the approximate ore particle size, and if the mean of the approximate ore particle size exceeds the preset ore particle size threshold range, an abnormal alarm of the grading screen is issued. That is, when the mean of the ore particle size is greater than the preset range of 8-13mm, it is determined that the grading screen is damaged or abnormal, and an abnormal alarm of the grading screen is issued.

[0046] The crushing detection of the oversize material is carried out, and the large pieces of raw coal after crushing are selected for secondary crushing. When the raw coal that needs secondary crushing exceeds the preset range, the alarm information of the crusher abnormality is sent; the crushing detection of the oversize material is carried out, and the large pieces of raw coal after crushing are selected for secondary crushing. When the raw coal that needs secondary crushing exceeds the preset range, the alarm information of the crusher abnormality is sent. The content specifically includes: using the U-NET network to extract the contour features of the large pieces of oversize material to obtain the first contour image; using the Canny algorithm to perform edge enhancement processing on the first contour image to obtain the second contour image; calculating the large pieces of raw coal with an area block greater than the preset threshold in the second contour image, and putting them into the crusher for secondary crushing; if the area of ​​the large pieces of raw coal in the second contour image is greater than the preset threshold range, the alarm information of the crusher abnormality is sent. The content of extracting the contour features of the large pieces of oversize material using the U-NET network to obtain the first contour image specifically includes: when using U-NET, adding an attention mechanism to extract features at the jump connection of the U-Net model; using the loss function to optimize the feature extraction content to obtain the first contour image.

[0047] The secondary crushed raw coal and the undersize are conveyed to the coal washer for classification to obtain gangue and medium coal; the secondary crushed raw coal and the undersize are conveyed to the coal washer for classification to obtain gangue and medium coal, and the content specifically includes: the gangue includes pure gangue and the first medium coal, and its density is greater than 1.6g / cm 3 The medium coal includes the second medium coal and clean coal, and its density is less than 1.6g / cm 3 .

[0048] The medium coal is subjected to secondary screening through the jig to obtain clean coal with smaller particle size. When the particle size of the clean coal screened twice does not meet the preset requirements, the coal after the secondary screening is put into the jig for tertiary screening. When the third screening still does not meet the preset requirements, an alarm message of jig abnormality is sent; the medium coal is subjected to secondary screening through the jig to obtain clean coal with smaller particle size. When the particle size of the clean coal screened twice does not meet the preset requirements, the coal after the secondary screening is put into the jig for tertiary screening. When the third screening still does not meet the preset requirements, an alarm message of jig abnormality is sent. The content of the alarm message of jig abnormality specifically includes: detecting the size of the clean coal. If the size meets the preset requirements, it is determined that there is no abnormality. If the size of the clean coal exceeds the preset requirements, secondary screening is performed. If the size of the clean coal after the secondary screening still does not meet the preset requirements, it is determined that the tool is abnormal, the jig operation is stopped, and a fault signal is sent.

[0049] In this embodiment, the density of clean coal is required to be 1.5 g / cm 3 .

[0050] Measure the clean coal that meets the preset requirements.

[0051] When measuring, it is necessary to calculate the weight of raw coal input within a certain period of time and the weight of clean coal obtained after the raw coal passes through the coal washing system.

[0052] In this embodiment, the time period of raw coal input is recorded, and the time when clean coal falls after the raw coal passes through the coal washing system is obtained according to the time required for the coal washing system to process the coal mine and the conveying rate of the conveyor belt. The weight is measured from the time when the clean coal falls, and the amount of raw coal converted into clean coal is calculated to obtain the clean coal yield. When the conversion rate is lower than the normal amount, it is detected whether each component of the coal washing system is used normally. If each component is abnormal, an abnormal alarm is sent.

[0053] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. An online detection and metering method applied to a coal washing system, characterized in that: The method comprises: The raw coal in the coal receiving machine is transferred to the grading screen, and the particle size of the material on the grading screen and the material under the grading screen is detected. When the particle size of the material under the grading screen is uneven, an alarm message indicating the abnormality of the grading screen is sent; The crushing detection is carried out on the screen material, and the large pieces of raw coal after crushing are selected for secondary crushing. When the raw coal that needs secondary crushing exceeds the preset range, an alarm message of crusher abnormality is sent; The secondary crushed raw coal and the undersize are sent to the coal washer for classification to obtain gangue and medium coal; The medium coal is screened twice by the jig to obtain clean coal with a smaller particle size. When the particle size of the clean coal screened twice does not meet the preset requirements, the coal after the secondary screening is put into the jig for a third screening. When the third screening still does not meet the preset requirements, an alarm message of abnormal jig is sent; Measure the clean coal that meets the preset requirements.

2. The online detection and metering method for coal washing system according to claim 1 is characterized in that: The raw coal in the coal receiving machine is transferred to the grading screen, and the particle size of the sieve material and the sieve material of the grading screen is detected. When the particle size of the sieve material is uneven, the alarm information of the grading screen abnormality is sent. The specific contents include: Extract the contour features of the raw coal and generate a cutting plan; The cut raw coal is conveyed to the grading screen through the coal receiving machine; The coal under the grading screen is tested for coal mine particle size using lifting wavelet and LBF model. When the test does not meet the particle size requirements, an abnormal alarm for the grading screen is sent.

3. The online detection and metering method for coal washing system according to claim 2 is characterized in that: The method for automatically generating a cutting plan specifically includes: Use superpixel algorithm to segment coal mine images and generate superpixel blocks; Merging the super pixel blocks using a merging strategy, and scoring the coal block image to be processed according to the region cutting; Labeling data of some of the coal block images to be processed, and constructing a segmentation model based on semi-supervised learning self-training according to the labeled data and the remaining data of the coal block images to be processed; The segmentation model based on semi-supervised learning and self-training is used to cut the coal mine image to achieve the cutting of the coal mine.

4. The online detection and metering method for coal washing system according to claim 1 is characterized in that: The coal under the grading screen is tested for coal mine particle size using lifting wavelet and LBF model. When the test does not meet the particle size requirements, the content of the abnormal alarm of the grading screen is sent, including: The cubic B-spline function is used as the prediction operator of lifting wavelet to filter and remove noise and detect edges of the acquired coal block image to be processed; Perform morphological operations on the wavelet transformed image to eliminate the holes in the coal mine image. The LBF model is used to cut the image after wavelet processing, and the connected city operation is performed on the cut image to obtain the number of coal mines. The number of pixels in the image is converted into the actual ore area, and the horizontal projection of the ore is equivalent to a circle. The approximate ore particle size is obtained through the area of ​​the circle. The average value of the approximate ore particle size is obtained. If the average value of the approximate ore particle size exceeds the preset ore particle size threshold range, an abnormal alarm of the grading screen is issued.

5. The online detection and metering method for coal washing system according to claim 1 is characterized in that: The crushing detection of the material on the screen is carried out, and the large pieces of raw coal after crushing are selected for secondary crushing. When the raw coal that needs secondary crushing exceeds the preset range, the alarm information of the crusher abnormality is sent, and the content specifically includes: The U-NET network is used to extract the contour features of the large pieces of material on the sieve to obtain the first contour image; Performing edge enhancement processing on the first contour image using a Canny algorithm to obtain a second contour image; Calculate the large pieces of raw coal whose area blocks are larger than a preset threshold in the second contour map and put them into the crusher for secondary crushing; If the area of ​​the large piece of raw coal in the second contour image is larger than the preset threshold range, an alarm message indicating the crusher is abnormal is sent.

6. The online detection and metering method for coal washing system according to claim 5 is characterized in that: The U-NET network is used to extract the contour features of the large pieces of material on the sieve, and the content of the first contour image obtained specifically includes: When using U-NET, add an attention mechanism to the jump connection of the U-Net model to extract features; The loss function is used to optimize the feature extraction content and obtain the first contour image.

7. The online detection and metering method for coal washing system according to claim 1 is characterized in that: The process of conveying the secondary crushed raw coal and the undersize to the coal washer for classification to obtain gangue and medium coal specifically includes: The gangue includes pure gangue and first medium coal, and its density is greater than 1.6g / cm 3 ; The medium coal includes the second medium coal and clean coal, and its density is less than 1.6g / cm 3 .

8. The on-line detection and metering method for coal washing system according to claim 1 is characterized in that: The medium coal is screened twice by the jig to obtain clean coal with smaller particle size. When the particle size of the clean coal screened twice does not meet the preset requirements, the coal after the secondary screening is put into the jig for the third screening. When the third screening still does not meet the preset requirements, the alarm information of jig abnormality is sent, including: The size of the clean coal is tested. If the size meets the preset requirements, it is judged that there is no abnormality. If the size of the clean coal exceeds the preset requirements, secondary jigging screening is carried out. If the size of the clean coal after the secondary jigging screening still does not meet the preset requirements, it is judged that the tool is abnormal, the jig machine operation is stopped, and a fault signal is issued.

Images