X-ray ash welt detection process

Abstract

The invention relates to the technical field of ash content detection, in particular to an X-ray ash content welt detection process. The process comprises the following steps: collecting a sample and crushing the sample to a preset particle size; shaping the crushed sample; performing shape detection on the shaped sample, and judging whether the shaped sample meets a detection standard or not; the sample meeting the detection standard is conveyed to the position below a detector window for fluorescence detection, the detector window is parallel to the top face of the sample, the distance between the detector window and the top face of the sample is 0-5 mm, and the repetition error precision of the obtained ash content detection result is within + / -0.4%. According to the method, the granularity and the form of the sample are limited by crushing, shaping and detecting the shape of the sample, so that the distance between the window of the detector and the top surface of the sample can be controlled to be 0-5mm, the repeated error precision of the obtained ash content detection result is controlled to be within + / -0.4%, and the precision and the stability of the detection result are improved.

Description

Technical Field

[0001] The invention relates to the technical field of ash content detection, and in particular to an X-ray ash content edge detection process. Background Art

[0002] In the fields of mineral processing, coal preparation, thermal power and coal chemical industry, it is necessary to detect the content of main components of raw materials or products. Traditionally, it is mainly detected by chemical titration (ore processing) or ash burning (coal). The detection cycle is long and there is a large delay in the guidance of production process parameter adjustment. Therefore, the online rapid detection technology of the main component content can effectively guide production and improve production quality.

[0003] At present, X-ray detection technology is mainly used for coal ash detection, which detects coal ash content by emitting X-rays through X-ray ash instruments. Most of the current online component detection systems use direct detection, that is, setting up X-ray ash instruments above the production input or output belts, and online calibration of materials of different particle sizes and costs, but the detection accuracy after calibration is not ideal. Summary of the invention The purpose of the present invention is to provide an X-ray ash edge detection process to solve the technical problem of low coal ash detection accuracy in the prior art.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions: An X-ray ash content edge detection process comprises the following steps: Collect samples and crush them to a preset particle size; Reshape the crushed samples; Perform morphological testing on the reshaped samples to determine whether the reshaped samples meet the testing standards; The samples that meet the detection standards are transported to the bottom of the detector window for fluorescence detection to obtain spectral characteristics, and the ash content detection results of the samples are calculated based on the spectral characteristics, wherein the detector window is parallel to the top surface of the sample, and the distance between the detector window and the top surface of the sample is 0~5mm, and the repeatability error accuracy of the ash content detection results is within ±0.4%.

[0005] Furthermore, the preset particle size is less than or equal to 3 mm.

[0006] Furthermore, the distance between the detector window and the top surface of the sample is 2 mm ± 0.15 mm, so that the repeatability error accuracy is within ± 0.25%.

[0007] Furthermore, the step of shaping the crushed sample specifically includes: The sample is shaped by a shaping device to shape the top surface of the sample into a plane, and the thickness of the shaped sample stream is at least 2 cm and the length is at least 30 cm.

[0008] Furthermore, the cross-section of the sample material flow is trapezoidal, with an upper cross-section width of 3 cm and a lower cross-section width of 7 cm.

[0009] Furthermore, the shaping device includes a shaping roller matched with the detection belt, and the step of shaping the sample using the shaping device specifically includes: The crushed sample is sent to the detection belt, and the moving belt speed of the detection belt is 5mm / s±0.2mm / s; At least one shaping roller is arranged above the detection belt along its moving direction, and the sample passes through the at least one shaping roller in sequence under the conveyance of the detection belt.

[0010] Furthermore, a substantially trapezoidal or rectangular shaping opening is formed between the detection belt and the shaping roller, and the shaping roller is provided in a plurality, so that a plurality of the shaping openings are formed above the detection belt, and the areas of the plurality of shaping openings tend to decrease along the conveying direction of the detection belt; And / or, the shaping roller is an active roller, and the rotation speed of the shaping roller is 4r / s; And / or, a brush is arranged above the shaping roller, and the brush sweeps away the sample stuck on the shaping roller during the rotation of the shaping roller.

[0011] Furthermore, the step of performing morphological detection on the shaped sample specifically includes: The sample material flow stably passes through the morphology detection camera at a preset moving belt speed, and the morphology detection camera detects at least one feature of the width, height and top hole area of ​​the sample.

[0012] Furthermore, the step of the morphology detection camera detecting at least one feature of the width, height and top hole area of ​​the sample comprises: Acquire an image of the sample and perform recognition analysis on the image and / or detect the height of the sample; Samples that do not meet the testing standards include: the width error of the sample is greater than or equal to ±5mm; and / or the height error of the sample is greater than or equal to ±1mm; and / or there are holes on the top surface of the sample with a diameter greater than or equal to 2mm; and / or there are missing parts in the sample and the area of ​​the missing parts accounts for greater than or equal to 3%.

[0013] Furthermore, the step of transporting the sample meeting the detection standard to the bottom of the detector window for detection specifically includes: The sample is tested using a fluorescence detection device, which includes a main structure, a support arm, an X-ray fluorescence instrument and a protective baffle, wherein: The upper end of the support arm is rotatably connected to the external frame, the lower end of the support arm is rotatably connected to the main structure, and the length of the support arm can be adjusted. The main structure is installed above the detection belt through the support arm; The X-ray fluorescence instrument is installed in the main structure, and a through hole is provided on the main structure, and the detector of the X-ray fluorescence instrument detects the sample on the detection belt through the through hole; The protective baffle is in a V-shaped structure and is installed on the side of the main structure where the material flow comes in. The tip surface of the protective baffle is used to scrape off samples that exceed the standard layer height.

[0014] Furthermore, it also includes: A spacing monitoring model is established according to the mapping relationship between the spacing between the detector window and the top surface of the sample and the spectral characteristics; The obtained spectral characteristics of the sample are input into the spacing monitoring model to calculate the current spacing between the detector window and the top surface of the sample.

[0015] Furthermore, the fluorescence detection adopts a rolling judgment method, and a judgment is performed every preset time period, and each judgment includes the following steps: In a preset time period, the detector is controlled to collect spectral data once at a set time interval, and a total of n spectral data are collected, where n is greater than or equal to 2; Performing normalization processing on each piece of the spectral data to obtain n pieces of relative spectral data; Using cluster screening to divide the n pieces of relative spectral data into cluster one and cluster two; Cluster one and cluster two are calculated to obtain spectral characteristics with coal flow and / or spectral characteristics without coal flow.

[0016] Furthermore, after the step of dividing the n pieces of relative spectrum data into cluster one and cluster two by cluster screening, each judgment further includes the following steps: Determine whether the distance between the center of cluster one and the center of cluster two is less than the preset center distance. If so, merge cluster one and cluster two into a single cluster. And / or, spectral data with intensities exceeding a preset range in cluster 1 and cluster 2 are respectively eliminated.

[0017] Furthermore, it also includes: the spectral characteristics include spectral characteristics with coal flow and spectral characteristics without coal flow, and the intensity change amplitude of the set element in the spectral characteristics without coal flow obtained each time is compared. If the intensity change amplitude of the set element exceeds a set threshold, it is determined that the detector window is dirty, wherein the set element is the element contained in the sample.

[0018] Beneficial effects of the present invention: The X-ray ash edge detection process provided by the present invention includes the following steps: collecting samples and crushing the samples to a preset particle size; shaping the crushed samples; performing morphological detection on the shaped samples to determine whether the shaped samples meet the detection standards; and conveying the samples that meet the detection standards to the bottom of the detector window for fluorescence detection, wherein the detector window is parallel to the top surface of the sample, and the distance between the detector window and the top surface of the sample is 0-5mm, and the repetitive error accuracy of the obtained ash detection results is within ±0.4%. The X-ray ash edge detection process provided by the present application limits the particle size and morphology of the samples by crushing, shaping and morphologically detecting the samples, so that the distance between the detector window and the top surface of the sample can be controlled between 0-5mm, so that the repetitive error accuracy of the obtained ash detection results is controlled within ±0.4%, effectively improving the accuracy and stability of the detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A schematic diagram of the process flow of an X-ray ash content edge detection process provided by an embodiment of the present invention; Figure 2 A schematic diagram of the structure of an X-ray ash content edge detection system provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a shaping device provided in an embodiment of the present invention; Figure 4 A schematic diagram of the movement of the shaping roller and the detection belt provided in an embodiment of the present invention; Figure 5 A schematic diagram of the structure of a morphology detection device provided by an embodiment of the present invention; Figure 6 A schematic diagram of the structure of a sample that meets the detection standard in an embodiment of the present invention; Figure 7 The structure of the sample that does not meet the detection standard in the embodiment of the present invention is shown in FIG. Figure 1 ; Figure 8 The structure of the sample that does not meet the detection standard in the embodiment of the present invention is shown in FIG. Figure 2 ; Fig. 9A schematic diagram of the structure of a fluorescence detection device provided in an embodiment of the present invention; Fig.10 A schematic structural diagram of a discarding device provided in an embodiment of the present invention.

[0021] icon: 1-sampling and crushing device; 11-primary processing module; 111-primary full-section sampling head; 112-primary crusher; 113-belt feeder; 114-primary automatic reducer; 12-secondary processing module; 121-secondary sampling belt; 122-secondary full-section sampling head; 123-secondary automatic reducer; 124-dryer; 125-secondary crusher; 126-vibrating screen; 127-bucket elevator; 2-cooling device; 21-constant volume silo; 22-fabricating mechanism; 3-detection belt; 4-shaping device; 41-shaping roller; 42-shaping mouth; 43-brush; 44-shaping roller driving source; 5-morphology detection device; 51-camera; 52-distance sensor; 53-light source; 6-discarding device; 61-linear driving source; 62-roller brush; 63-waste chute; 7-fluorescence detection device; 71-main structure; 72-protective baffle; 73-support arm; 8-moisture detection device; 9-temperature detection device. DETAILED DESCRIPTION

[0022] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, rather than 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.

[0023] It should be noted that in the description of the present invention, the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance.

[0024] As mentioned in the background technology, most of the current online component detection systems use a direct detection method, that is, an X-ray ash instrument is set up above the production input or output belt to perform online calibration on materials of different particle sizes and costs, but the detection accuracy after calibration is not ideal.

[0025] Based on this, the present invention provides an X-ray ash edge detection process, referring to Figure 1, the process comprises the following steps: Collect samples and crush them to a preset particle size; Reshape the crushed samples; Perform morphological testing on the reshaped samples to determine whether the reshaped samples meet the testing standards; The samples that meet the detection standards are transported to the bottom of the detector window for fluorescence detection to obtain spectral characteristics, and the ash content detection results of the samples are calculated based on the spectral characteristics, wherein the detector window is parallel to the top surface of the sample, and the distance between the detector window and the top surface of the sample is 0~5mm, and the repeatability error accuracy of the ash content detection results is within ±0.4%.

[0026] During the detection process, the detector collects spectral data at a set time interval (for example, every 1 second), and obtains the final spectrum diagram by processing the collected spectral data; the spectrum diagram contains a variety of spectral features, such as common element characteristics (Si, Ca, Ti, Fe, etc.) in coal samples and scattering characteristics of the spectral background, etc. The ash content detection value of the sample can be obtained by calculating various element characteristics. In order to make the detection results more accurate, the fluorescence detection provided in this application adopts a rolling judgment method, and a judgment is made every preset time period (for example, 2 minutes); in each judgment, the detector collects a total of 120 spectral data at a frequency of collecting spectral data once per second, and obtains a more accurate spectrum diagram by screening and averaging the 120 spectral data.

[0027] The X-ray ash content edge detection process provided in the present application shapes the crushed sample into a preset shape, thereby avoiding the problem of irregular sample surface. The shaped sample must also undergo a morphological detection step to ensure that the sample's morphology meets the requirements of the preset morphology. Since the particle size and morphology of the sample are limited, the distance between the detector window and the top surface of the sample can be controlled between 0 and 5 mm, so that the repeatability error accuracy of the ash content detection result is controlled within ±0.4%, effectively improving the detection accuracy. In addition, the reduction in detection distance also reduces the difficulty of protection, thereby making the working environment safer.

[0028] Optionally, the distance between the detector window and the top surface of the sample can be 0 mm, 1 mm, 2 mm, 3 mm, 4 mm, or 5 mm.

[0029] Furthermore, the step of collecting samples and crushing the samples to a preset particle size specifically includes: collecting samples from the main belt, performing a primary crushing process on the collected samples, crushing the samples to less than 13 mm and reducing them to form primary samples; collecting the reduced primary samples, performing a secondary crushing process on the primary samples, and controlling the particle size of the finally formed samples to be less than or equal to 3 mm. Based on the above steps, the primary samples are dried before the secondary crushing process is performed on the primary samples.

[0030] In a specific embodiment, the step of collecting samples and crushing the samples to a preset particle size includes: collecting 10kg~30kg of samples from the main belt, performing primary crushing treatment on the collected samples, crushing the samples to less than 13mm and reducing them to 5kg±0.4kg to form primary samples; collecting 3kg±0.2kg of primary samples, and drying the collected primary samples for 180 seconds using a microwave of more than 10kw and hot air at 85℃; performing secondary closed-circuit crushing on the dried primary samples, and sieving the materials after the secondary crushing treatment through a vibrating screen so that the particle size of the final sample is less than or equal to 3mm.

[0031] When the particle size of the sample is greater than 3mm, the coal sample segregates severely and it is difficult to ensure the repeatability accuracy. Therefore, the process provided in this application controls the particle size of the sample to within 3mm through a two-stage crushing process (it can be all particle sizes within 3mm, such as 0.5mm, 1mm, 2mm, 2.3mm, 3mm), which can ensure that the repeatability error accuracy is controlled within ±0.4%. If the preset particle size exceeds 3mm, the coal sample will be severely segregated. At this time, even if the distance between the detector window and the top surface of the sample is 0, the repeatability error accuracy of the ash detection result is greater than ±0.4%, which ultimately leads to poor detection accuracy. In addition, the ash content corresponding to different coal qualities is different, which means that if you want the repeatability error accuracy to always be ±0.4%, you need to conduct precise analysis to determine the distance between the detector window and the top surface of the sample.

[0032] After experimental verification, when the particle size of the sample is less than or equal to 3mm, for the case where the distance between the detector window and the top surface of the sample is 1mm, the standard deviation of the ash content detected by the X-ray fluorescence instrument is 0.1288, and the repeatability error precision is 0.12%; for the case where the distance between the detector window and the top surface of the sample is 3mm, the standard deviation of the ash content detected by the X-ray fluorescence instrument is 0.2452, and the repeatability error precision is 0.18%; for the case where the distance between the detector window and the top surface of the sample is 5mm, the standard deviation of the ash content detected by the X-ray fluorescence instrument is 0.4852, and the repeatability error precision is 0.35%. It can be seen that the process provided by the present application can ensure the repeatability error precision within ±0.4% by controlling the particle size of the sample within 3mm and the distance between the detector window and the top surface of the sample within 5mm, thereby improving the accuracy and stability of the detection results.

[0033] Preferably, on the basis that the particle size of the sample is less than or equal to 3 mm, the distance between the detector window and the top surface of the sample is 2 mm±0.15 mm, so that the repeatability error accuracy is within ±0.25%.

[0034] Furthermore, after the step of collecting samples and crushing the samples to a preset particle size, the X-ray ash edge detection process further includes the steps of quantifying and cooling the samples to 24°C.

[0035] Specifically, the X-ray ash edge detection process provided in this application is implemented using an X-ray ash edge detection system. Figure 2 The X-ray ash edge detection system includes a sampling crushing device 1, a detection belt 3, a shaping device 4, a morphological detection device 5 and a fluorescence detection device 7, wherein: the sampling crushing device 1 is used to collect samples and crush the samples; the detection belt 3 is used to receive the samples output from the sampling crushing device 1 and sequentially convey the samples to the shaping device 4, the morphological detection device 5 and the fluorescence detection device 7; the shaping device 4 is used to shape the samples into a set shape; the morphological detection device 5 is used to perform morphological detection on the shaped samples; and the fluorescence detection device 7 is used to perform fluorescence detection on the samples. A fixed volume silo 21 is arranged between the sampling crushing device 1 and the detection belt 3, and the crushed samples are sent into the fixed volume silo 21, and the samples exceeding the maximum capacity of the fixed volume silo 21 overflow, so that the samples can be quantitatively measured; the samples entering the fixed volume silo 21 are quickly cooled by the heat dissipation mechanism on the fixed volume silo 21, so that the samples are cooled to 24°C, so as to avoid the sample temperature being too high and affecting the detection effect of the detector; the samples after heat dissipation fall vertically from the outlet at the bottom of the fixed volume silo 21 onto the detection belt 3.

[0036] In the above structure, the capacity of the constant-volume silo 21 can be adjusted according to actual conditions. For example, the capacity of the constant-volume silo 21 can be 2L. The distance between the bottom surface of the outlet of the constant-volume silo 21 and the detection belt 3 can be 4 cm. This distance allows the sample to smoothly fall onto the detection belt 3 and also prevents blockage of the material at the outlet position of the constant-volume silo 21.

[0037] Continue to refer to Figure 1 and Figure 2 The step of shaping the crushed sample specifically includes: using the shaping device 4 to shape the sample so as to make the top surface of the sample flat, and the thickness of the shaped sample material flow is at least 2 cm and the length is at least 30 cm.

[0038] Refer to Figure 2 and Figure 3 The shaping device 4 includes a shaping roller 41 that cooperates with the detection belt 3. On this basis, the step of using the shaping device 4 to shape the sample specifically includes: Sending the crushed sample onto the detection belt 3, and the moving belt speed of the detection belt 3 is 5 mm / s ± 0.2 mm / s; At least one shaping roller 41 is arranged above the detection belt 3 along its moving direction. The sample passes through at least one shaping roller 41 in sequence under the conveyance of the detection belt 3 to form a sample material flow with a thickness of at least 2 cm and a length of at least 30 cm.

[0039] Optionally, a shaping opening 42 that is generally trapezoidal or rectangular is formed between the detection belt 3 and the shaping roller 41. The number of shaping rollers 41 is multiple, so that a plurality of shaping openings 42 are formed above the detection belt 3, and the areas of the plurality of shaping openings 42 show a decreasing trend along the conveyance direction of the detection belt 3; And / or, the shaping roller 41 is a driving roller, and the rotation speed of the shaping roller 41 is 4 r / s; And / or, a brush 43 is arranged above the shaping roller 41, and the brush 43 sweeps away the sample adhered to the shaping roller 41 during the rotation of the shaping roller 41.

[0040] In this embodiment, the crushed sample first enters the constant-volume silo 21 for quantitative measurement and cooling, and the cooled sample falls onto the detection belt 3. Two shaping rollers 41 are sequentially arranged above the detection belt 3 along its moving direction, so that two shaping openings 42 are formed above the detection belt 3. The shaping roller 41 is generally a V-shaped roller, and the shaping opening 42 formed by the shaping roller 41 and the detection belt 3 is generally trapezoidal. Each shaping roller 41 is driven by an independent shaping roller drive source 44 (such as a motor), and its rotation speed is 4 r / s. A brush 43 is arranged above each shaping roller 41 to sweep away the material adhered to the shaping roller 41.

[0041] In the above structure, since the shaping port 42 is roughly trapezoidal, the sample passing through the shaping port 42 is also shaped into a sample material flow with a trapezoidal cross section; compared with a rectangular sample material flow, the structure of the trapezoidal sample material flow is more stable and not easy to collapse. Each shaping roller 41 is driven by an independent shaping roller driving source 44, such as Figure 4 As shown, the instantaneous movement direction of the contact point between the shaping roller 41 and the detection belt 3 is opposite to the movement direction of the detection belt 3, so that the surface of the sample flow can be wiped backwards, making the surface of the sample smoother; in addition, when the shaping roller 41 passes through the brush 43, the material contaminated on the shaping roller 41 is blocked by the brush 43 and brushed to the rear side of the sample flow, preventing the brushed material from falling onto the sample flow that has been shaped, thereby further improving the flatness of the top surface of the sample.

[0042] In this embodiment, the length of the middle straight cylinder of the first shaping roller 41 is 6 cm and the diameter is 4 cm, the length of the two side cones is 3 cm, the diameter of the small end of the cone is 4 cm and the diameter of the large end is 9 cm; the length of the middle straight cylinder of the second shaping roller 41 is 3 cm and its diameter is 5 cm, the length of the two side cones is 2 cm, the diameter of the small end of the cone is 5 cm and the diameter of the large end is 9 cm. The above structure makes the areas of the two shaping ports 42 decrease successively along the conveying direction. When the sample passes through the first shaping port 42, the sample is shaped into a trapezoidal surface with a cross-sectional thickness of 2.5 cm, an upper cross-sectional width of 6 cm, and a lower cross-sectional width of 12 cm. When the sample passes through the second shaping port 42, the sample is shaped into a trapezoidal surface with a cross-sectional thickness of 2 cm, an upper cross-sectional width of 3 cm, and a lower cross-sectional width of 7 cm (such as Figure 6 As shown in FIG. 1 , the length of the sample material flow is at least 30 cm. Since at least two shaping rollers 41 are used to shape the sample, the material flow is gradually shaped into a preset shape, which can reduce the difficulty of shaping the material flow and ensure the shaping effect.

[0043] Further, continue to refer to Figure 1 and Figure 2 The step of performing morphological detection on the shaped sample specifically includes: the sample material flow stably passes through a morphological detection camera at a preset moving belt speed, and the morphological detection camera detects at least one feature of the sample's width, height, and top hole area.

[0044] The step of the morphology detection camera detecting at least one feature of the width, height and top hole area of ​​the sample comprises: Acquire an image of the sample and perform recognition analysis on the image and / or detect the height of the sample; Samples that do not meet the testing standards include: the width error of the sample is greater than or equal to ±5mm; and / or the height error of the sample is greater than or equal to ±1mm; and / or there are holes with a diameter greater than or equal to 2mm on the top surface of the sample; and / or there are missing parts in the sample and the area of ​​the top surface of the missing parts accounts for greater than or equal to 3%.

[0045] This embodiment uses a morphology detection device 5 to perform morphology detection on the sample, and the morphology detection device 5 includes a camera 51 mounted above the detection belt 3. When the sample flow passes through the camera 51 under the conveyance of the detection belt 3, the camera 51 can detect at least one feature of the width, height and top hole area of ​​the sample.

[0046] In this embodiment, the preset shape of the sample is a trapezoidal surface with a cross-sectional thickness of 2 cm, an upper cross-sectional width of 3 cm, and a lower cross-sectional width of 7 cm, and a sample flow length of at least 30 cm. For samples that meet any of the following requirements: the width error of the sample is greater than or equal to ±5 mm, the height error of the sample is greater than or equal to ±1 mm, and there are holes with a diameter greater than or equal to 2 mm on the top surface of the sample (such as Figure 7 As shown) and the sample has missing parts and the area of ​​the missing parts is greater than or equal to 3% (as shown Figure 8 As shown in the figure, all samples were judged as not meeting the test standards.

[0047] On the basis of the above steps, after the step of performing morphological inspection on the shaped samples, the X-ray ash edge inspection process further includes: pushing the samples that do not meet the inspection standards laterally off the inspection belt 3.

[0048] Furthermore, before or after the step of conveying the sample meeting the inspection standard to the bottom of the detector window for inspection, the X-ray ash edge inspection process further includes: performing moisture inspection and / or temperature inspection on the sample meeting the inspection standard.

[0049] To realize the above process, the X-ray ash edge detection system also includes a moisture detection device 8 and a temperature detection device 9. The sample that meets the detection standard is transported by the detection belt 3 and stably moves through the fluorescence detection device 7, the moisture detection device 8 and the temperature detection device 9 at the belt speed. The moisture detection device 8 can be a capacitive contact detector, a near-infrared non-contact detector, or a microwave detector, which is used to measure the moisture content of the sample material flow. The temperature detection device 9 can be a contact or non-contact temperature sensor, which is used to measure the temperature of the sample material flow.

[0050] Reference Fig. 9In this embodiment, the fluorescence detection device 7 includes a main structure 71, an X-ray fluorescence instrument, a protective baffle 72 and a distance sensor; the main structure 71 is installed above the detection belt 3, and the X-ray fluorescence instrument is installed in the main structure 71; the protective baffle 72 is a V-shaped structure, and is installed on the side of the main structure 71 where the material flow comes in, and the tip surface of the protective baffle 72 is used to scrape off the sample that exceeds the standard layer height to prevent damage to the detection probe of the X-ray fluorescence instrument; the distance sensor is installed on the side of the main structure 71 where the material flow comes in, and is used to detect the thickness of the sample material flow, and further ensure that the height of the sample material flow meets the requirements. The main structure 71 has a bottom surface facing the detection belt 3, and a concave structure is provided on the bottom surface, and a through hole is opened in the concave structure. The X-ray fluorescence instrument detects the sample on the detection belt 3 through the through hole, so that the distance between the X-ray fluorescence instrument and the belt surface of the detection belt 3 is 2cm, and the distance between the detector window of the X-ray fluorescence instrument and the top surface of the sample is 0~5mm, and the preferred distance is 2mm±0.15mm. In addition, the fluorescence detection device 7 also includes a support arm 73 for connecting to the external frame. The main structure is rotatably connected to the support arm 73, and the length of the support arm 73 is adjustable, so that the distance between the detection probe of the X-ray fluorescence instrument and the sample is always kept constant, thereby improving the detection accuracy of the X-ray fluorescence instrument.

[0051] Optionally, the distance sensor in the fluorescence detection device 7 is an ultrasonic rangefinder. The moisture detection device 8 is a near-infrared moisture detector, and the near-infrared moisture detector is set up just above the sample flow, 10 cm away from the top surface of the flow. The temperature detection device 9 is specifically a PT100 temperature sensor, which is a rod-shaped PT100 temperature sensor with a diameter of 1 cm and a length of 10 cm. It is perpendicular to the top surface of the flow and is inserted into the flow from the upper middle part of the flow, with an insertion depth of 1 cm. X-ray fluorescence instrument, ultrasonic rangefinder, near-infrared moisture detector and temperature sensor are all mature instruments, and the principle is not repeated here.

[0052] Finally, the sample is analyzed for composition. After the sample that meets the test standards has been tested for temperature, the continuously collected data such as X-ray fluorescence spectrum, moisture, temperature, etc. are processed to obtain the concentration of various elements.

[0053] In summary, an X-ray ash content edge detection process provided by a specific embodiment of the present application includes the following steps: S1: Collect samples and perform primary crushing, drying and secondary crushing on the samples in sequence, reduce the samples to 3kg±0.2kg, and control the particle size of the samples to be less than or equal to 3mm; S2: quantify and cool the crushed samples to 24°C; S3: shaping the cooled sample into a trapezoidal shape with a cross-sectional thickness of 2 cm, an upper cross-sectional width of 3 cm, a lower cross-sectional width of 7 cm, and a sample stream with a length of at least 30 cm; S4: Perform morphological testing on the shaped sample to determine whether the shaped sample meets the testing standard; Samples that meet the testing standards are transported to the next process; For samples that do not meet the test standards, they are pushed off the test belt 3 laterally; S5: transport the sample that meets the test standard to the bottom of the detector window of the X-ray fluorescence instrument for fluorescence detection, wherein the detector window is parallel to the top surface of the sample, and the distance between the detector window and the top surface of the sample is 2mm±0.15mm, and the repeatability error accuracy of the ash content test result is within ±0.25%; S6: Conduct moisture testing on samples that meet the testing standards; S7: Conduct temperature testing on samples that meet the testing standards; S8: Analyze the composition of the sample and process the continuously collected X-ray fluorescence spectrum, moisture, temperature and other data to obtain the concentration of various elements.

[0054] Furthermore, the X-ray ash edge detection process provided in the present application also includes the following steps: establishing a spacing monitoring model based on the mapping relationship between the spacing between the detector window and the top surface of the sample and the spectral characteristics; inputting the obtained spectral characteristics of the sample into the spacing monitoring model to calculate the current spacing between the detector window and the top surface of the sample.

[0055] It is found in actual detection that the distance between the detector window and the top surface of the sample mainly affects the background characteristics (background shape and background intensity) in the spectrum graph. Therefore, the above-mentioned spectral characteristics can be specifically background intensity. For example, before actual detection, a distance monitoring model is established based on the attenuation relationship between the distance between the detector window and the top surface of the sample and the background intensity value (i.e., the integral area within the specified spectral channel range); during the actual detection process, the current distance between the detector window and the top surface of the coal flow sample is inferred through the background intensity in the real-time collected spectral characteristics. In one embodiment, the step of establishing the spacing monitoring model specifically includes: When the distance between the detector window and the top surface of the coal flow sample is 0, h1, h2...hn, the background intensity values ​​of multiple groups (e.g., 5 groups) of coal flow samples (e.g., the sum of the spectral intensity in the range of 1000-1500 channels) are measured respectively, and the average value of the background intensity values ​​of the 5 groups of samples is calculated; wherein, 0, h1, h2...hn increases in sequence, for example, h1=1 mm, h2=2 mm, and so on, hn=n mm; considering the influence of the acquisition time and current, the background intensity value can be normalized; Using the experimentally measured data, the relationship between the distance between the detector window and the top surface of the sample and the background intensity value (i.e., the distance monitoring model) is fitted. For example, the relationship is: h=a+b*back, where h is the distance between the detector window and the top surface of the sample, back is the background intensity value, and a and b are mapping coefficients calculated based on the experimentally measured data.

[0056] In the actual detection process, the background intensity value measured each time or at intervals is substituted into the above formula to obtain the current distance between the detector window and the top surface of the sample, so that the staff can understand the actual detection situation and adjust the distance between the detector window and the top surface of the sample in time to ensure that the distance meets the detection requirements. In order to remind the staff to make adjustments in time, the X-ray ash edge detection system can be controlled to issue an alarm when the distance between the detector window and the top surface of the sample is greater than the preset value (for example, 1 mm).

[0057] Furthermore, the fluorescence detection adopts a rolling judgment method, and a judgment is performed every preset time period, and each judgment includes the following steps: In a preset time period, the detector is controlled to collect spectral data once at a set time interval, and a total of n spectral data are collected, where n is greater than or equal to 2; In order to avoid the influence of dimension on clustering effect, each piece of spectral data is normalized (such as Z-score normalization) to obtain n pieces of relative spectral data. The calculation formula is as follows:

[0058] Where: is the spectral data collected each time; Using clustering screening (such as k-means or DBSCAN density clustering algorithm) to divide the n pieces of relative spectral data into cluster one and cluster two; Cluster one and cluster two are calculated to obtain spectral characteristics with coal flow and / or spectral characteristics without coal flow.

[0059] It should be noted that, in the case where part of the spectral data detected by the detector during the preset time period is spectral data without coal flow and part of it is spectral data with coal flow, cluster one is spectral data without coal flow and cluster two is spectral data with coal flow. At this time, the corresponding spectral intensity of cluster one is calculated and averaged to obtain the final spectral map without coal flow, and the corresponding spectral intensity of cluster two is calculated and averaged to obtain the final spectral map with coal flow. In the case where all the spectral data detected by the detector during the preset time period are spectral data without coal flow, cluster one and cluster two are all spectral data without coal flow. At this time, cluster one and cluster two are merged into a single cluster, and the corresponding spectral intensity of the single cluster is calculated and averaged to obtain the final spectral map without coal flow. Similarly, in the case where all the spectral data detected by the detector during the preset time period are spectral data with coal flow, cluster one and cluster two are all spectral data with coal flow. At this time, cluster one and cluster two are merged into a single cluster, and the corresponding spectral intensity of the spectral data in the single cluster is calculated and averaged to obtain the final spectral map with coal flow.

[0060] This embodiment uses the K-means clustering algorithm (also known as the K-means clustering algorithm) to divide the n pieces of relative spectral data into cluster 1 and cluster 2. The K-means clustering algorithm is an iterative clustering analysis algorithm. Each time an object is assigned, the cluster center (or cluster center) of the cluster is recalculated based on the existing objects in the cluster. The cluster center after each iteration is the mean of the existing objects in the cluster in each dimension.

[0061] Exemplarily, the step of using K-means clustering screening to divide 120 relative spectral data into cluster one and cluster two specifically includes: right Sort and select the initial cluster center with the smallest value as cluster one , select the initial cluster center with the maximum value as cluster 2 , the calculation formula is: =min( ), =max( ) In each iteration, each data point Assign to the closest cluster:

[0062] Where: k is the number of iterations; is the cluster center of cluster one after each iteration, = , is the total number of data in cluster 1, is the sum of all data in cluster 1; is the cluster center of cluster 2 after each iteration, = , is the total number of data in cluster 2, is the sum of all data in cluster 2.

[0063] As an optional embodiment, after the step of dividing the n pieces of relative spectrum data into cluster one and cluster two by cluster screening, each judgment further includes the following steps: Determine whether the distance between the cluster center of cluster 1 and the cluster center of cluster 2 is less than the preset center distance. The calculation formula is: <

[0064] Where: is the cluster center of cluster 1; is the cluster center of cluster 2; The preset center distance is selected according to the data situation, usually 3 times the average value of the standard deviation of the two clusters; If yes, it means that all the spectral data in this time period are spectral data without coal flow or spectral data with coal flow, then cluster one and cluster two are merged into a single cluster.

[0065] As an optional embodiment, after the step of using cluster screening to divide the n relative spectral data into cluster one and cluster two, each judgment also includes the following steps: respectively judging whether the number of data in cluster one and cluster two is less than a preset number (for example, less than 10% of the total number of data); if the number of data in either cluster one or cluster two is less than the preset number, merging cluster one and cluster two into a single cluster.

[0066] As an optional embodiment, after the step of using cluster screening to divide the n pieces of relative spectral data into cluster one and cluster two, each judgment further includes the following steps: respectively eliminating the spectral data in cluster one and cluster two whose intensity exceeds the preset range. For cluster two (i.e., spectral data with coal flow), the spectral data with too small intensity (which may be caused by too small coal flow or not close to the equipment) is directly discarded without subsequent processing. For cluster one (i.e., spectral data without coal flow), the spectral data with too large intensity (which may be caused by some coal flow residue in empty detection) is directly discarded without subsequent processing.

[0067] The step of removing the spectral data whose intensity exceeds the preset range in cluster 1 and cluster 2 respectively specifically includes: calculating the standard deviation σ1 of cluster 1 and the standard deviation σ2 of cluster 2 respectively, and the calculation formula is:

[0068] Where: j = 1, 2; is the total number of samples in the jth cluster; For the data values; is the sample mean of the jth cluster; Eliminate excess ±1.5 The spectral data of is taken as outliers. After removing the outliers, the stable concentrated data is obtained as the final cluster spectral data.

[0069] In summary, the fluorescence detection provided in this embodiment adopts a rolling judgment method, and a judgment is performed every preset time period to ensure that no detection is missed. In each judgment, the K-means clustering algorithm is first used to divide the collected spectral data into two clusters; if the distance between the cluster centers of the two clusters is less than the preset center distance, it means that both clusters of data are spectral data without coal flow or spectral data with coal flow, then the two clusters are merged into one cluster for calculation to obtain a spectral graph without coal flow or a spectral graph with coal flow; if the distance between the cluster centers of the two clusters is greater than the preset center distance, then the two clusters of data are spectral data without coal flow and spectral data with coal flow, respectively, and the two clusters are calculated separately to obtain spectral graphs without coal flow and spectral graphs with coal flow, respectively. In addition, before calculating cluster one, the spectral data with excessive intensity in cluster one is eliminated, and before calculating cluster two, the spectral data with excessive intensity in cluster two is eliminated to obtain stable and concentrated data as the final cluster spectral data. Furthermore, the X-ray ash edge detection process provided in the present application also includes: the spectral characteristics include spectral characteristics with coal flow and spectral characteristics without coal flow, and the intensity change amplitude of the set element (such as Si) in the spectral characteristics without coal flow obtained each time is calculated. If the intensity change amplitude exceeds the set threshold, it is determined that the detector window is dirty; wherein, the set element is the element contained in the sample.

[0070] During the production process, various problems may occur in the equipment, among which the detector window becoming dirty is one of the common situations. There are two main reasons for the detector window to become dirty: one is that it becomes dirty slowly, usually caused by the accumulation of fly ash during the production process; the other is that it suddenly sticks to coal. Both of these situations will cause the coal element characteristics in the spectral characteristics of the coal-free flow to change. For example, when the detector window is not contaminated with a little coal sample, the intensity of the silicon element in the spectral characteristics of the coal-free flow is the initial intensity value; as more and more coal samples are placed on the detector window, the intensity of the silicon element in the spectral characteristics of the coal-free flow is also affected by the coal samples on the detector window. Increasingly, therefore, the coal characteristics in the spectral characteristics of the coal-free flow can be analyzed to determine whether the beryllium window is dirty. The specific operation is to extract the intensity of the silicon element in the spectral characteristics of the coal-free flow and compare it with the historical data (which can be the above-mentioned initial intensity value); if the intensity change amplitude of the silicon element exceeds the set threshold, it is determined that the detector window is dirty to a certain extent. Once it is detected that the beryllium window is judged to be dirty to a certain extent, the system will immediately issue an alarm signal to remind the staff to deal with it in time. At the same time, the system can also automatically issue an instruction to clean the detector window in order to quickly restore the normal working state of the equipment and ensure the accuracy of subsequent spectral acquisition.

[0071] To sum up, the X-ray ash content edge detection process that this application wants to protect is the original creation of our R&D team after many years of research and development. Neither the types of factors affecting the accuracy of repeated errors nor the corresponding parameters can be obtained through simple experiments, so this technology is in a leading position in the industry; therefore, the innovative X-ray ash content edge detection process uses the above process to make the sample form a sample of specified particle size, morphology, humidity and temperature, so that the detector can perform close-range detection of the sample, and in the detection process, the spectral characteristics of the sample obtained each time are used to reversely infer the current distance between the detector window and the top surface of the sample, thereby ensuring that the distance between the detector window and the top surface of the sample meets the detection requirements, and the repeatability error accuracy of the test results of the sample component content is controlled within ±0.25%, which greatly improves the accuracy and stability of the test results.

[0072] To realize the above process, the present invention also provides an X-ray ash edge detection system, referring to Figure 2 The X-ray ash edge detection system includes a sampling and crushing device 1, a detection belt 3, a shaping device 4, a morphology detection device 5 and a fluorescence detection device 7, wherein: the sampling and crushing device 1 is used to collect samples and crush the samples; the detection belt 3 is used to receive the samples output from the sampling and crushing device 1 and transport the samples to the shaping device 4, the morphology detection device 5 and the fluorescence detection device 7 in sequence; the shaping device 4 is used to shape the sample into a set shape; the morphology detection device 5 includes a camera 51 and / or a ranging sensor 52 mounted above the detection belt 3; the fluorescence detection device 7 is used to perform fluorescence detection on the sample.

[0073] Furthermore, the sampling and crushing device 1 comprises a primary processing module 11 for performing primary crushing processing on the sample and a secondary processing module 12 for performing secondary crushing processing on the sample, wherein: The primary processing module 11 includes a primary full-section sampling head 111, a primary crusher 112, a belt feeder 113 and a primary automatic reducer 114 which are arranged in sequence; And / or, the secondary processing module 12 is sequentially provided with a secondary sampling belt 121, a secondary full-section sampling head 122, a secondary automatic reducer 123, a dryer 124, a secondary crusher 125 and a vibrating screen 126.

[0074] The working process of the above-mentioned primary processing module 11 is as follows: when sampling starts, the first-level full-section sampling head 111 collects samples from the main belt according to the full-section of the material flow, and the collected samples enter the first-level crusher 112 along the chute, and the first-level crusher 112 crushes the samples to less than 13mm; after the samples are discharged from the first-level crusher 112, they fall into the belt feeder 113, and the belt feeder 113 sends the samples to the first-level automatic reducer 114, which reduces the samples to 5kg±0.4kg to form primary samples.

[0075] In this embodiment, the secondary processing module 12 also includes a bucket elevator 127 arranged between the secondary crusher 125 and the vibrating screen 126. The working process of the above-mentioned secondary processing module 12 is as follows: the primary sample after being reduced by the primary automatic reducer 114 enters the secondary sampling belt 121, and the secondary full-section sampling head 122 collects samples from the secondary sampling belt 121 according to the full section of the material flow. The sample enters the secondary automatic reducer 123 and is reduced to 3kg±0.2kg of material, and then enters the dryer 124 for drying, and is fed into the secondary crusher 125. After crushing, the material falls into the vibrating screen 126 and is screened according to the standard particle size of 3mm; the fine particle sample particles screened by the vibrating screen 126 fall into the next process, and the coarse particle sample on the vibrating screen 126 falls into the bucket elevator 127 and returns to the secondary crusher 125 for crushing again.

[0076] Optionally, the number of the primary crusher 112 and the secondary crusher 125 can be one or more, and the specific number is set based on the requirement of the discharge particle size. The above crusher can be a cone crusher, a double-roll crusher, a jaw crusher, a double-roll crusher or other types of crushers, and can be used in a single unit or a combination of multiple types to achieve the discharge particle size requirement.

[0077] Optionally, the dryer 124 can be a microwave hot air dryer, which dries the sample by combining microwave drying and hot air drying to ensure that the material remains dry. In addition, the dryer 124 can also be a vacuum freezing device.

[0078] Furthermore, the X-ray ash edge detection system also includes a cleaning mechanism. Before the secondary processing module 12 starts sampling each time, the secondary processing module 12 is first purged by the cleaning mechanism using high-pressure air. Specifically, the cleaning mechanism includes a linear cylinder and a high-pressure air pipe, and the linear cylinder drives the high-pressure air pipe to reciprocate to spray the inside of the dryer 124, the secondary crusher 125, the vibrating screen 126, and the bucket elevator 127 to blow away the contaminated material powder, and at the same time start the dust removal fan to discharge the dust to keep the secondary processing module 12 from being contaminated by the previous batch of materials. In a specific embodiment, the purging pressure is 0.5~1MPa, and the blowing time is 40 seconds.

[0079] Furthermore, the X-ray ash edge detection system also includes a cooling device 2 arranged between the sampling and crushing device 1 and the detection belt 3, and the cooling device 2 includes a constant volume silo 21, a heat dissipation mechanism, a temperature sensor and a level meter, wherein: the constant volume silo 21 is used to receive the sample output from the sampling and crushing device 1 and convey the sample to the detection belt 3; the heat dissipation mechanism is used to dissipate the heat inside the constant volume silo 21; the temperature sensor is used to detect the internal temperature of the constant volume silo 21; the level meter is used to detect the material level of the constant volume silo 21.

[0080] Specifically, the heat dissipation mechanism includes at least one heat dissipation pipe and a heat dissipation fan, wherein: the heat dissipation pipe is inserted into the constant volume silo 21 horizontally; the heat dissipation fan can blow air into the constant volume silo 21 to dissipate heat. Taking the number of heat dissipation pipes as eight as an example, every four heat dissipation pipes form a layer, and the four heat dissipation pipes in each layer are distributed at 90° along the center of the constant volume silo 21, and the spacing between the upper and lower layers of heat dissipation pipes is 5 cm. Fins are provided on the periphery of the heat dissipation pipe to increase the contact area with the cold air and improve the heat dissipation effect. A plurality of (for example, three) temperature sensors are evenly arranged on the outer circumference of the silo wall of the constant volume silo 21, and at least one temperature sensor is arranged on the upper and lower sides of the center of the silo of the constant volume silo 21. The material level meter is installed on the top of the constant volume silo 21 to detect whether the constant volume silo 21 is full of material.

[0081] Furthermore, the cooling device 2 further includes a material distribution mechanism 22, which is located below the constant volume silo 21 and includes an electrically controlled gate, a flexible chute and at least one pair of squeezing rollers, wherein: The electric control gate is arranged at the outlet of the constant volume silo 21; The flexible chute extends from the outlet of the constant volume silo 21 to the detection belt 3; Each pair of squeezing rollers includes two squeezing rollers respectively arranged on both sides of the flexible chute.

[0082] The working principle of the material distributing mechanism 22 is as follows: the electrically controlled gate opens to discharge the material after the fixed volume silo 21 is full, and the material falls vertically into the flexible chute from the outlet of the fixed volume silo 21; during the discharging process, the squeezing roller moves up and down along the flexible chute and squeezes the material in the flexible chute, and the deformation of the flexible chute prevents the material inside from being blocked.

[0083] In other embodiments, the combination structure of the flexible chute and the squeezing roller may be replaced by a combination structure of a rigid chute and a vibration motor, which can also prevent the material inside the chute from being blocked.

[0084] Reference Figure 2 and Figure 3 The shaping device 4 includes at least one shaping roller 41 disposed above the detection belt 3. When there are multiple shaping rollers 41, the multiple shaping rollers 41 are arranged in sequence along the moving direction of the detection belt 3. The sample passes through at least one shaping roller 41 in sequence under the conveyance of the detection belt 3, forming a sample material flow with a thickness of at least 2 cm and a length of at least 30 cm.

[0085] On the basis of the above structure, a substantially trapezoidal or rectangular shaping opening 42 is formed between the detection belt 3 and the shaping roller 41; there are multiple shaping rollers 41, so that multiple shaping openings 42 are formed above the detection belt 3, and the areas of the multiple shaping openings 42 tend to decrease along the conveying direction of the detection belt 3; and / or, the shaping device 4 further comprises a shaping roller driving source 44 for driving the shaping roller 41 to rotate; And / or, the shaping device 4 further includes a brush 43 arranged above the shaping roller 41 .

[0086] In this embodiment, two shaping rollers 41 are sequentially arranged above the detection belt 3 along its moving direction, thereby forming two shaping openings 42 above the detection belt 3. The shaping roller 41 is roughly V-shaped, and the shaping opening 42 surrounded by the shaping roller 41 and the detection belt 3 is roughly trapezoidal. Each shaping roller 41 is driven by an independent shaping roller driving source 44 (such as a motor); and a brush 43 is arranged above each shaping roller 41 to sweep away the material stuck on the shaping roller 41.

[0087] In other embodiments, the shaping roller 41 may also be a driven roller, or the shaping roller 41 may also be replaced by a trowel plate.

[0088] Reference Figure 5, the morphology detection device 5 includes a camera 51, a distance sensor 52 and a light source 53, wherein: the camera 51 and the distance sensor 52 are vertically mounted above the detection belt 3, and the distance between the two and the sample flow after shaping is approximately 30 cm; the number of light sources 53 can be three, which are respectively arranged on the top and both sides of the detection belt 3 to illuminate the top and both sides of the sample flow. Optionally, the light sources 53 on both sides are at an angle of 45° to the moisture surface. It should be noted that the camera 51, the distance sensor 52 and the light source 53 are not limited to the above settings. For example, the camera 51 can also obliquely shoot the image of the sample flow on the detection belt 3; the number of light sources 53 can also be one or more.

[0089] Reference Fig. 9 The fluorescence detection device 7 includes a main structure 71, an X-ray fluorescence instrument and a protective baffle 72, wherein: the main structure 71 is installed above the detection belt 3; the X-ray fluorescence instrument is installed in the main structure 71, and a through hole is provided on the main structure 71, through which the detector of the X-ray fluorescence instrument detects the sample on the detection belt 3; the protective baffle 72 is a V-shaped structure, and is installed on the side of the main structure 71 where the material flow comes in, and the tip surface of the protective baffle 72 is used to scrape off the sample exceeding the standard layer height to prevent damage to the detection probe of the X-ray fluorescence instrument.

[0090] In order to further control the thickness of the sample material flow passing through the main structure 71, the fluorescence detection device 7 can also include a distance measuring sensor, which is installed on the side of the main structure 71 where the material flow comes in, and is used to detect the thickness of the sample material flow.

[0091] On the basis of the above structure, a concave structure is provided on the bottom surface of the main structure 71, and a through hole is opened in the concave structure. The X-ray fluorescence instrument detects the sample on the detection belt 3 through the through hole in the concave structure.

[0092] When the above-mentioned fluorescence detection device 7 is in use, when the sample passes through the main structure 71 under the conveyance of the detection belt 3, the protective baffle 72 can push away the sample on the detection belt 3 that exceeds the standard layer height; because the height of the sample passing through the X-ray fluorescence instrument is strictly controlled, the distance between the detector window of the X-ray fluorescence instrument and the sample on the detection belt 3 can be shortened. In addition, the bottom surface of the main structure 71 can further smooth the sample on the detection belt 3, so that the detection accuracy can be further improved. A recessed structure is provided on the bottom surface of the main structure 71, and a through hole is opened in the recessed structure, so that a gap is generated between the recessed structure and the surface of the sample, thereby preventing the sample from entering the main structure 71 through the through hole on the recessed structure and damaging the detector.

[0093] Furthermore, the fluorescence detection device 7 also includes a support arm 73, the upper end of the support arm 73 is rotatably connected to the external frame, the lower end of the support arm 73 is rotatably connected to the main structure 71, and the main structure 71 is installed above the detection belt 3 through the support arm 73. Since the support arm 73 is rotatably connected to the external frame and the main structure 71, it can play a certain buffering role when the detection belt 3 passes, avoiding damage to the X-ray fluorescence instrument inside the main structure 71. In addition, the length of the support arm 73 is adjustable, so that the distance between the detection probe of the X-ray fluorescence instrument and the sample can be adjusted and the distance can be kept constant at all times, thereby improving the detection accuracy of the X-ray fluorescence instrument.

[0094] Continue to refer to Figure 2 The X-ray ash content detection also includes a discard device 6 arranged between the morphology detection device 5 and the fluorescence detection device 7, and the discard device 6 is used to push the samples that do not meet the detection standards laterally off the detection belt 3.

[0095] Specifically, refer to Fig.10 The discarding device 6 includes a linear drive source 61, a roller brush 62 and a waste chute 63, wherein: the linear drive source 61 is used to drive the roller brush 62 to move laterally above the detection belt 3; the waste chute 63 is arranged below one side of the detection belt 3, so that the roller brush 62 pushes the sample on the detection belt 3 into the waste chute 63 during the lateral movement. Optionally, the linear drive source 61 is specifically a cylinder. The cylinder body of the cylinder is fixed to one side of the detection belt 3, and the waste chute 63 is located on the side of the detection belt 3 away from the cylinder body of the cylinder. The roller brush 62 is connected to the end of the piston rod of the cylinder through a connecting frame.

[0096] On the basis of the above structure, the discarding device 6 further includes a rotating driving source (not shown in the figure), which may be a motor, the body of which is fixedly mounted on the connecting frame, and the output shaft of which is connected to the roller brush 62, so as to drive the roller brush 62 to rotate during the lateral movement. Fig. 9 As shown, during the lateral movement of the roller brush 62, the rotary driving source drives the roller brush 62 to rotate clockwise to clean the sample on the detection belt 3. In a specific embodiment, the rotation speed of the roller brush 62 is 20r / s, and the discarding device 6 is arranged 30cm downstream of the morphology detection device 5.

[0097] When the material flow detection fails, the roller brush 62 and the baffle plate are pushed out by the linear drive source 61 and move horizontally above the detection belt 3, while the roller brush 62 rotates to clean the material, and the baffle plate guides the waste material to the waste chute 63 for discharge. The roller brush 62 can reciprocate multiple times to clean the residual material to the waste chute 63.

[0098] In summary, the system for implementing the X-ray ash edge detection process provided in this embodiment includes a sampling and crushing device 1, a cooling device 2, a detection belt 3, a shaping device 4, a morphology detection device 5, a discarding device 6, a fluorescence detection device 7, a moisture detection device 8 and a temperature detection device 9, wherein: the sampling and crushing device 1 is used to collect samples and crush the samples; the discharge port of the sampling and crushing device 1 is located above the feed port of the cooling device 2, and the discharge port of the cooling device 2 is located above the detection belt 3; along the conveying direction of the detection belt 3, the cooling device 2, the shaping device 4, the morphology detection device 5, the discarding device 6, the fluorescence detection device 7, the moisture detection device 8 and the temperature detection device 9 are arranged in sequence. Among them, the shaping device 4, the morphology detection device 5 and the discarding device 6 can ensure that the sample conveyed to the fluorescence detection device 7 meets the requirements of particle size, morphology, humidity and temperature, so that the detector can perform close-range (0~5mm) detection on the sample, and the repeatability error accuracy of the detection result of the sample component content is controlled within ±0.4%, which greatly improves the accuracy and stability of the detection result.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements 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. An X-ray ash content edge detection process, characterized in that: The following steps are involved: Collect samples and crush them to a preset particle size; Reshape the crushed samples; Perform morphological testing on the reshaped samples to determine whether the reshaped samples meet the testing standards; The sample meeting the detection standard is transported to the bottom of the detector window for fluorescence detection to obtain the spectral characteristics, and the ash content detection result of the sample is calculated according to the spectral characteristics, wherein the detector window is parallel to the top surface of the sample, and the distance between the detector window and the top surface of the sample is 0-5mm, and the repeatability error accuracy of the obtained ash content detection result is within ±0.4%; A spacing monitoring model is established based on the mapping relationship between the spacing between the detector window and the top surface of the sample and the spectral characteristics; the spectral characteristics of the obtained sample are input into the spacing monitoring model to calculate the current spacing between the detector window and the top surface of the sample.

2. The X-ray ash content edge detection process according to claim 1, characterized in that: The preset particle size is less than or equal to 3 mm.

3. The X-ray ash content edge detection process according to claim 2, characterized in that: The distance between the detector window and the top surface of the sample is 2 mm ± 0.15 mm, so that the repeatability error accuracy is within ± 0.25%.

4. The X-ray ash content edge detection process according to claim 1, characterized in that: The step of shaping the crushed sample specifically comprises: The sample is shaped by a shaping device to shape the top surface of the sample into a plane, and the thickness of the shaped sample stream is at least 2 cm and the length is at least 30 cm.

5. The X-ray ash content edge detection process according to claim 4, characterized in that: The cross section of the sample stream is trapezoidal, with an upper cross section width of 3 cm and a lower cross section width of 7 cm.

6. The X-ray ash content edge detection process according to claim 4, characterized in that: The shaping device includes a shaping roller matched with the detection belt, and the step of shaping the sample by using the shaping device specifically includes: The crushed sample is sent to the detection belt, and the moving belt speed of the detection belt is 5mm / s±0.2mm / s; At least one shaping roller is arranged above the detection belt along its moving direction, and the sample passes through the at least one shaping roller in sequence under the conveyance of the detection belt.

7. The X-ray ash content edge detection process according to claim 6, characterized in that: A substantially trapezoidal or rectangular shaping opening is formed between the detection belt and the shaping roller, and the shaping roller is in a plurality, so that a plurality of shaping openings are formed above the detection belt, and the areas of the plurality of shaping openings tend to decrease along the conveying direction of the detection belt; And / or, the shaping roller is an active roller, and the rotation speed of the shaping roller is 4r / s; And / or, a brush is arranged above the shaping roller, and the brush sweeps away the sample stuck on the shaping roller during the rotation of the shaping roller.

8. The X-ray ash content edge detection process according to claim 1, characterized in that: The step of performing morphological detection on the shaped sample specifically comprises: The sample material flow stably passes through the morphology detection camera at a preset moving belt speed, and the morphology detection camera detects at least one feature of the width, height and top hole area of ​​the sample.

9. The X-ray ash content edge detection process according to claim 8, characterized in that: The step of the morphology detection camera detecting at least one feature of the width, height and top hole area of ​​the sample comprises: Acquire an image of the sample and perform recognition analysis on the image and / or detect the height of the sample; Samples that do not meet the testing standards include: the width error of the sample is greater than or equal to ±5mm; and / or the height error of the sample is greater than or equal to ±1mm; and / or there are holes on the top surface of the sample with a diameter greater than or equal to 2mm; and / or there are missing parts in the sample and the area of ​​the missing parts accounts for greater than or equal to 3%.

10. The X-ray ash content edge detection process according to claim 1, characterized in that: The step of transporting the sample meeting the detection standard to the bottom of the detector window for detection specifically includes: The sample is tested using a fluorescence detection device, which includes a main structure, a support arm, an X-ray fluorescence instrument and a protective baffle, wherein: The upper end of the support arm is rotatably connected to the external frame, the lower end of the support arm is rotatably connected to the main structure, and the length of the support arm can be adjusted. The main structure is installed above the detection belt through the support arm; The X-ray fluorescence instrument is installed in the main structure, and a through hole is provided on the main structure, and the detector of the X-ray fluorescence instrument detects the sample on the detection belt through the through hole; The protective baffle is in a V-shaped structure and is installed on the side of the main structure where the material flow comes in. The tip surface of the protective baffle is used to scrape off samples that exceed the standard layer height.

11. The X-ray ash edge detection process according to any one of claims 1 to 10, characterized in that: The fluorescence detection adopts a rolling judgment method, and a judgment is performed every preset time period. Each judgment includes the following steps: In a preset time period, the detector is controlled to collect spectral data once at a set time interval, and a total of n spectral data are collected, where n is greater than or equal to 2; Performing normalization processing on each piece of the spectral data to obtain n pieces of relative spectral data; Using cluster screening to divide the n pieces of relative spectral data into cluster one and cluster two; Cluster one and cluster two are calculated to obtain spectral characteristics with coal flow and / or spectral characteristics without coal flow.

12. The X-ray ash content edge detection process according to claim 11, characterized in that: After the step of dividing the n pieces of relative spectrum data into cluster one and cluster two by cluster screening, each judgment further includes the following steps: Determine whether the distance between the cluster center of cluster one and the cluster center of cluster two is less than the preset center distance. If so, merge cluster one and cluster two into a single cluster. and / or, determining whether the number of data in cluster 1 and cluster 2 is less than a preset number respectively, and if the number of data in any one of cluster 1 and cluster 2 is less than the preset number, merging cluster 1 and cluster 2 into a single cluster; And / or, spectral data with intensities exceeding a preset range in cluster 1 and cluster 2 are respectively eliminated.

13. The X-ray ash content edge detection process according to any one of claims 1 to 10, characterized in that: Also includes: The spectral characteristics include spectral characteristics with coal flow and spectral characteristics without coal flow. The intensity change amplitude of the set element in the spectral characteristics without coal flow obtained each time is calculated. If the intensity change amplitude exceeds the set threshold, it is determined that the detector window is dirty; wherein the set element is the element contained in the sample.

Images