Mineral composition online detection system, method and device

Abstract

The invention provides a mineral composition online detection system, method and device, and the system comprises an energy spectrum collection subsystem which comprises an X-ray tube, a high-voltage controller and an energy spectrum detector; the high-voltage controller controls the X-ray tube to emit X-rays to the to-be-detected mineral; when the energy spectrum detector obtains a trigger signal that the to-be-detected mineral reaches the detection position, spectrum data of the to-be-detected mineral is collected, and an initial energy spectrum signal of the to-be-detected mineral is obtained; the information acquisition subsystem comprises a laser ranging device; the laser ranging device determines morphology information of the surface of the to-be-detected mineral; and the data processing subsystem corrects the initial energy spectrum signal based on the morphology information to obtain a target energy spectrum signal. According to the scheme of the invention, the interference information of the initial energy spectrum signal is eliminated, and an accurate substance composition result is obtained.

Description

Technical Field

[0001] The present application relates to the field of X-ray technology, and particularly to an on-line detection system, method and device for mineral components. Background Art

[0002] X-rays are electromagnetic radiation with a wavelength of 0.01 - 10 nanometers and a frequency of 30 - 30000 PHz. That is, X-rays are electromagnetic waves with extremely high frequency, extremely short wavelength and large energy. X-rays have penetrability. When X-rays interact with matter, X-ray fluorescence (XRF) is generated. By measuring and analyzing the energy spectrum signal of X-ray fluorescence, the elemental composition of the matter can be known and the material components can be obtained.

[0003] However, due to the interference of the actual environment, there is interference information in the energy spectrum signal of X-ray fluorescence. When determining the material components based on the energy spectrum signal of X-ray fluorescence, incorrect results may be obtained. Summary of the Invention

[0004] The present application provides an on-line detection system for mineral components, which at least includes:

[0005] A sample preparation subsystem for performing on-line sample preparation operations on the mineral to be detected;

[0006] An energy spectrum acquisition subsystem located downstream of the sample preparation subsystem. The energy spectrum acquisition subsystem at least includes an X-ray tube, a high-voltage controller and an energy spectrum detector. The high-voltage controller is used to control the X-ray tube to emit X-rays to the mineral to be detected. The energy spectrum detector is used to collect the spectral data of the mineral to be detected and obtain the initial energy spectrum signal of the mineral to be detected when a trigger signal indicating that the mineral to be detected reaches the detection position is obtained. Among them, the initial energy spectrum signal includes the X-ray fluorescence signal and the scattered light signal generated after the mineral to be detected is irradiated by the X-rays;

[0007] An information acquisition subsystem, which includes a laser ranging device. The laser ranging device is used to collect the topography information of the surface of the mineral to be detected;

[0008] A data processing subsystem for obtaining the initial energy spectrum signal and the topography information, where the topography information is used to characterize the roughness and / or flatness of the surface of the mineral to be detected. Based on the topography information, the initial energy spectrum signal is corrected to obtain a corrected target energy spectrum signal;

[0009] Among them, the target energy spectrum signal is used to analyze the material components of the mineral to be detected.

[0010] The present application provides an on-line detection method for mineral components, and the method includes:

[0011] An X-ray is emitted to the mineral to be detected, and the X-ray fluorescence and scattered light generated after the mineral to be detected is irradiated by the X-ray are detected to obtain the initial energy spectrum signal of the mineral to be detected;

[0012] Obtain the topography information of the surface of the mineral to be detected, where the topography information is used to characterize the roughness and / or flatness of the surface of the mineral to be detected;

[0013] Based on the topography information, the initial energy spectrum signal is corrected to obtain a corrected target energy spectrum signal, and the target energy spectrum signal is used to analyze the material composition of the mineral to be detected.

[0014] The present application provides an on-line detection device for mineral composition, and the device includes:

[0015] An acquisition module, configured to acquire an initial energy spectrum signal and the topography information of the surface of the mineral to be detected, where the topography information is used to characterize the roughness and / or flatness of the surface of the mineral to be detected; wherein, the initial energy spectrum signal is obtained by detecting the X-ray fluorescence and scattered light generated after the mineral to be detected is irradiated by the X-ray; a correction module, configured to correct the initial energy spectrum signal based on the topography information to obtain a corrected target energy spectrum signal, and the target energy spectrum signal is used to analyze the material composition of the mineral to be detected.

[0016] The present application provides an electronic device, including: a processor and a machine-readable storage medium, where the machine-readable storage medium stores machine-executable instructions that can be executed by the processor; the processor is configured to execute the machine-executable instructions to implement the on-line detection method for mineral composition in the above example of the present application.

[0017] The present application provides a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the on-line detection method for mineral composition in the above example is implemented.

[0018] The present application provides a machine-readable storage medium, where the machine-readable storage medium stores machine-executable instructions that can be executed by a processor; wherein, the processor is configured to execute the machine-executable instructions to implement the on-line detection method for mineral composition in the above example.

[0019] As can be seen from the above technical solutions, in the embodiments of the present application, after obtaining the initial energy spectrum signal of the X-ray fluorescence, the initial energy spectrum signal can be corrected to obtain a target energy spectrum signal. By correcting, the interference information in the initial energy spectrum signal is excluded, and then based on the target energy spectrum signal, the material composition of the mineral to be detected is analyzed, and an accurate material composition result can be obtained, avoiding obtaining an incorrect material composition result. Description of the Drawings

[0020] Figure 1 is a schematic structural diagram of the on-line mineral composition detection system in this application;

[0021] Figure 2 is a schematic diagram of the X-ray fluorescence generation process in this application;

[0022] Figure 3 is a schematic structural diagram of the X-ray generating device in this application;

[0023] Figure 4 is a schematic structural diagram of the on-line mineral composition detection system in this application;

[0024] Figure 5 is a schematic flow diagram of the on-line mineral composition detection method in this application;

[0025] Figure 6 is a schematic structural diagram of the on-line mineral composition detection device in this application. Detailed implementation manners

[0026] In an embodiment of this application, an on-line mineral composition detection system is proposed. Refer to Figure 1 shown, which is a schematic structural diagram of the on-line mineral composition detection system. The on-line mineral composition detection system may include:

[0027] A sample preparation subsystem for performing on-line sample preparation operations on the mineral to be detected;

[0028] An energy spectrum acquisition subsystem, located downstream of the sample preparation subsystem. The energy spectrum acquisition subsystem at least includes an X-ray tube, a high-voltage controller, and an energy spectrum detector; the high-voltage controller is used to control the X-ray tube to emit X-rays to the mineral to be detected; the energy spectrum detector is used to collect spectral data of the mineral to be detected when a trigger signal indicating that the mineral to be detected reaches the detection position is obtained, so as to obtain an initial energy spectrum signal of the mineral to be detected; the initial energy spectrum signal includes an X-ray fluorescence signal and a scattered light signal generated after the mineral to be detected is irradiated by X-rays;

[0029] An information acquisition subsystem, and the information acquisition subsystem includes a laser ranging device; the laser ranging device is used to collect the topography information of the surface of the mineral to be detected;

[0030] A data processing subsystem for obtaining the initial energy spectrum signal and the topography information of the surface of the mineral to be detected, where the topography information is used to characterize the roughness and / or flatness of the surface of the mineral to be detected; and correcting the initial energy spectrum signal based on the topography information to obtain a corrected target energy spectrum signal.

[0031] Among them, the target energy spectrum signal is used to analyze the material composition of the mineral to be detected.

[0032] Exemplarily, the topography information includes the heights of multiple position points on the surface of the mineral to be detected; when the data processing subsystem corrects the initial energy spectrum signal based on the topography information to obtain the corrected target energy spectrum signal, it is specifically used for: determining the target roughness based on the variance of the heights of the multiple position points;

[0033] Querying the obtained roughness correction information table based on the initial energy spectrum signal and the target roughness to obtain the target roughness correction coefficient, where the target roughness correction coefficient is used to characterize the influence degree of the roughness of the surface of the mineral to be detected on the energy spectrum signal; wherein, the roughness correction information table includes the corresponding relationship between the energy spectrum signal, the roughness and the roughness correction coefficient; correcting the initial energy spectrum signal based on the target roughness correction coefficient to obtain the corrected target energy spectrum signal.

[0034] Exemplarily, when the data processing subsystem corrects the initial energy spectrum signal based on the topography information to obtain the corrected target energy spectrum signal, it is specifically used for:

[0035] Determining the target height based on the average value of the heights of the multiple position points; querying the obtained height correction information table based on the initial energy spectrum signal and the target height to obtain the target height correction coefficient, where the target height correction coefficient is used to characterize the influence degree of the height of the surface of the mineral to be detected on the energy spectrum signal; wherein, the height correction information table includes the corresponding relationship between the energy spectrum signal, the height and the height correction coefficient; correcting the initial energy spectrum signal based on the target roughness correction coefficient and the target height correction coefficient to obtain the corrected target energy spectrum signal.

[0036] Exemplarily, the information acquisition subsystem further includes a temperature sensor, which is used to collect the current temperature information on the surface of the X-ray tube; the data processing subsystem is further used to obtain the current temperature information, determine the target temperature based on the current temperature information; query the obtained temperature correction information table based on the initial energy spectrum signal and the target temperature to obtain the target temperature correction coefficient, where the target temperature correction coefficient is used to characterize the influence degree of the temperature on the surface of the X-ray tube on the energy spectrum signal; wherein, the temperature correction information table includes the corresponding relationship between the energy spectrum signal, the temperature and the temperature correction coefficient; correcting the initial energy spectrum signal based on the target roughness correction coefficient and the target temperature correction coefficient to obtain the corrected target energy spectrum signal.

[0037] Exemplarily, if the target calibration information further includes the target humidity, the information acquisition subsystem further includes a humidity sensor for collecting the current humidity information in the X-ray path; the data processing subsystem is further configured to obtain the current humidity information, determine the target humidity based on the current humidity information; query the acquired humidity calibration information table based on the initial energy spectrum signal and the target humidity to obtain the target humidity calibration coefficient, where the target humidity calibration coefficient is used to characterize the influence degree of the humidity in the X-ray path of the X-ray tube on the energy spectrum signal; wherein, the humidity calibration information table includes the corresponding relationship between the energy spectrum signal, humidity and humidity calibration coefficient; and correct the initial energy spectrum signal based on the target roughness calibration coefficient and the target humidity calibration coefficient to obtain the calibrated target energy spectrum signal.

[0038] Exemplarily, the data processing subsystem is further configured to perform operations on the target height calibration coefficient, the target roughness calibration coefficient, the target temperature calibration coefficient and the target humidity calibration coefficient based on the relationship satisfied among the target height calibration coefficient, the target roughness calibration coefficient, the target temperature calibration coefficient and the target humidity calibration coefficient to obtain a comprehensive calibration coefficient; and correct the initial energy spectrum signal based on the comprehensive calibration coefficient to obtain the target energy spectrum signal.

[0039] Exemplarily, the sample preparation subsystem may include a detection belt, a scraper and a pressing roller; the detection belt is used for conveying the mineral to be detected; the scraper is used for leveling the mineral to be detected when the detection belt conveys the mineral to be detected to the scraper; the pressing roller is used for flattening the mineral to be detected when the detection belt conveys the mineral to be detected to the pressing roller; wherein, the on-line sample preparation operation includes a leveling operation and a flattening operation. For example, after the sample preparation subsystem flattens the mineral to be detected, the mineral to be detected passes through the energy spectrum acquisition subsystem and the information acquisition subsystem to complete data acquisition, and the acquired data is corrected to complete the component detection.

[0040] Exemplarily, the on-line mineral component detection system may further include an additional subsystem, and the additional subsystem may include but is not limited to at least one of the following: a temperature control unit, a radiation shielding protection device and a fan.

[0041] Wherein, the temperature control unit is used to control the temperature of the on-line mineral component detection system within a preset temperature range. The radiation shielding protection device is used to prevent the X-rays received by the energy spectrum detector from causing radiation to the outside.

[0042] The energy spectrum acquisition subsystem further includes a dust-proof component, and the dust-proof component is located between the X-ray tube and the mineral to be detected, and the X-ray passes through the dust-proof component to reach the surface of the mineral to be detected; for example, if the dust-proof component is a mylar film, the fan is used to blow air on the surface of the mylar film to prevent condensation on the surface of the mylar film.

[0043] Exemplarily, the data processing subsystem is further configured to, after obtaining the target energy spectrum signal, input the target energy spectrum signal into a trained neural network model (such as a dual-modal neural network inference model), so that the neural network model outputs signal features based on the target energy spectrum signal, and the signal features are used to analyze the material composition of the mineral to be detected. For example, the signal features represent the features of the material composition, and the material composition of the mineral to be detected can be directly determined based on the signal features. Alternatively, indirect calculations can also be performed based on the signal features, and the material composition of the mineral to be detected can be determined based on the indirect calculation results. There is no limitation in this regard, as long as the material composition of the mineral to be detected can be analyzed based on the signal features. Among them, the material composition may include, but is not limited to, one or more of ash composition, ash content, volatile matter, carbon and hydrogen, ash fusion point, total moisture, total sulfur, and calorific value.

[0044] Exemplarily, the on-line mineral composition detection system further includes a spectral subsystem (such as a near-infrared spectrometer). The spectral subsystem is configured to collect the near-infrared spectral signal of the mineral to be detected, and the data processing subsystem is configured to obtain the near-infrared spectral signal from the spectral subsystem. After obtaining the target energy spectrum signal and the near-infrared spectral signal, the target energy spectrum signal and the near-infrared spectral signal are input into a trained neural network model, so that the neural network model outputs signal features based on the target energy spectrum signal and the near-infrared spectral signal. For example, the neural network model fuses the target energy spectrum signal and the near-infrared spectral signal to obtain a fused signal, and determines and outputs signal features based on the fused signal. The signal features are used to analyze the material composition of the mineral to be detected.

[0045] As can be seen from the above technical solutions, in the embodiments of the present application, after obtaining the initial energy spectrum signal of X-ray fluorescence, the initial energy spectrum signal is corrected based on the target correction coefficient to obtain the target energy spectrum signal. The interference information in the initial energy spectrum signal is excluded through correction, and then the material composition of the mineral to be detected is analyzed based on the target energy spectrum signal, so as to obtain accurate material composition results and avoid obtaining incorrect material composition results.

[0046] The above technical solutions of the embodiments of the present application will be described below in conjunction with specific application scenarios.

[0047] X-rays are penetrative. When X-rays interact with matter, X-ray fluorescence is generated. By measuring and analyzing the energy spectrum signal of X-ray fluorescence, the elemental composition of the matter can be known and the material composition can be obtained. Based on this, the detection of the material composition is realized based on the principle that X-rays interact with matter atoms to generate fluorescence.

[0048] See Figure 2As shown, it is a schematic diagram of the X-ray fluorescence generation process. X-rays can interact with matter in three ways that affect X-ray detection: the photoelectric effect, Compton scattering (Compton scattering is incoherent scattering), and Rayleigh scattering (Rayleigh scattering is coherent scattering). The process of emitting X-rays can be called X-ray fluorescence, and the process of emitting X-rays can be simplified into two processes, namely scattering and fluorescence.

[0049] For scattering, the incident X-rays emitted from the X-ray tube, in Compton scattering, the incident photon interacts with the extranuclear electrons of the atom, transfers some energy to the electrons, and then generates a scattered photon and a high-energy electron. In addition, the incident X-rays emitted from the X-ray tube, in Rayleigh scattering, the incident photon changes direction, but does not lose energy after the interaction.

[0050] For X-ray fluorescence (photoelectric effect), the atoms in the sample are excited by the incident photons and collide with electrons, resulting in the removal of electrons from the inner layer or the generation of vacancies in the atom, which makes the atomic state unstable. The vacancies are filled by outer electrons, restoring the atom to a stable state. The transition from the outer electron orbit to the inner electron orbit is accompanied by the emission of secondary X-rays, which are recorded when they hit the detector. Each element has a unique set of energy levels, so experimental measurements can be carried out without destroying the elemental composition in the sample. The energy of the X-ray fluorescence photon is a property of the element and is equal to the energy difference between two electron energy levels.

[0051] The energy of the X-ray fluorescence photon provides qualitative information about the elemental characteristics. By measuring and analyzing the energy or wavelength of the X-rays, it is possible to know which element it is, so as to determine the substance composition and quantify the content of each element in the substance. The number or intensity of the X-ray fluorescence photons is characteristic of the number or concentration of the current element. The radiation intensity of each element signal is proportional to the concentration of the element in the sample, can be calculated through a set of calibration curves, and can be directly displayed in concentration units. The total count is expressed as the intensity in counts per second.

[0052] In the spectral analysis of X-ray fluorescence, both the K shell and the L shell will be detected. The X-ray spectrum irradiating the sample will show multiple peaks of different intensities, namely the fluorescence lines of the K, L, etc. shells. The energy of the X-rays generated by this reaction is discrete and is equal to the difference in the electron binding energies of these two atomic shells.

[0053] See Figure 3 As shown, it is a schematic diagram of the structure of the X-ray generating device. Regarding the process of generating X-rays by the X-ray generating device, this process can include: applying a high voltage to the cathode of the X-ray tube, accelerating the electrons generated by the hot filament and bombarding the target material, and the electrons interact with the target material to generate X-rays and emit them from the window.

[0054] Due to the interference of the actual environment, there is interference information in the energy spectrum signal of X-ray fluorescence. When determining the substance composition based on the energy spectrum signal of X-ray fluorescence, incorrect substance composition analysis results may be obtained.

[0055] For example, after the X-ray exits from the window, the X-ray is absorbed and attenuated by nitrogen, oxygen, and moisture in the air. Therefore, X-ray fluorescence is easily affected by the humidity in the X-ray optical path, and X-ray fluorescence is easily affected by the distance between the substance to be measured and the X-ray tube / detector. After the X-ray reaches the surface of the substance to be measured, it reacts with the substance to be measured to generate X-ray fluorescence. The emitted X-ray fluorescence is diffuse scattered light and is easily blocked by an uneven surface. Therefore, the rough detection surface of the substance to be measured will affect the intensity of X-ray fluorescence. The external environmental temperature affects the operating temperature of the X-ray tube, and the intensity of the X-ray is different at different operating temperatures.

[0056] In summary, it can be seen that there may be interference information caused by humidity, temperature, distance, and roughness in the energy spectrum signal of X-ray fluorescence. These interference information will all lead to incorrect substance composition analysis results based on the energy spectrum signal of X-ray fluorescence.

[0057] In view of the above findings, an on-line detection system for mineral composition is proposed in the embodiments of the present application, which can achieve high-robust correction of X-ray fluorescence signals. By correcting the influence of the environmental temperature on the X-ray fluorescence signal, the interference information caused by temperature in the energy spectrum signal of X-ray fluorescence is eliminated, and the measurement accuracy of the energy spectrum signal of X-ray fluorescence is improved. By correcting the influence of humidity on the X-ray fluorescence signal, the interference information caused by humidity in the energy spectrum signal of X-ray fluorescence is eliminated, and the measurement accuracy of the energy spectrum signal of X-ray fluorescence is improved. By correcting the influence of the height of the sample preparation surface on the X-ray fluorescence signal, the interference information caused by height in the energy spectrum signal of X-ray fluorescence is eliminated, and the measurement accuracy of the energy spectrum signal of X-ray fluorescence is improved. By correcting the influence of the roughness of the sample preparation surface on the X-ray fluorescence signal, the interference information caused by roughness in the energy spectrum signal of X-ray fluorescence is eliminated, and the measurement accuracy of the energy spectrum signal of X-ray fluorescence is improved.

[0058] An on-line detection system for mineral composition is proposed in the embodiments of the present application. Refer to Figure 4 As shown in the figure, it is a schematic structural diagram of the on-line detection system for mineral composition. The on-line detection system for mineral composition may include, but is not limited to: a sample preparation subsystem (also referred to as a sample preparation component), an energy spectrum acquisition subsystem (also referred to as an energy spectrum acquisition component), an information acquisition subsystem (also referred to as an information acquisition component), a data processing subsystem (also referred to as a data processing component), and an additional subsystem (also referred to as an additional component).

[0059] Exemplarily, the sample preparation subsystem is used to perform sample preparation operations on the placed minerals to be detected. Refer to Figure 4 As shown, the sample preparation subsystem may include, but is not limited to, a detection belt 1, a scraper 2, and a pressure roller 3.

[0060] The detection belt 1 can be a belt conveyor or other conveyor device, and can place the minerals to be detected on the detection belt 1, and the detection belt 1 conveys the minerals to be detected.

[0061] When the minerals to be detected are conveyed by the detection belt 1 to the scraper 2, the scraper 2 levels the minerals to be detected. When the minerals to be detected are conveyed by the detection belt 1 to the pressure roller 3, the pressure roller 3 flattens the minerals to be detected, and the flattened minerals to be detected are represented by 13 and subsequently denoted as minerals to be detected 13.

[0062] After the sample preparation subsystem flattens the minerals to be detected, the minerals to be detected complete data acquisition through the energy spectrum acquisition subsystem and the information acquisition subsystem, and the acquired data completes component detection after calibration.

[0063] For example, the minerals to be detected can also be called substances to be measured. For example, the minerals to be detected can be coal or other types of minerals to be detected. The scraper 2 levels the coal, and the pressure roller 3 flattens the coal, and the leveling operation and the flattening operation are sample preparation operations for the minerals to be detected.

[0064] Of course, here only the leveling operation and the flattening operation are taken as examples. For different types of minerals to be detected, different sample preparation operations can be adopted. That is, the scraper 2 and the pressure roller 3 are only examples, and devices adapted to the sample preparation operation can be selected to perform sample preparation operations on the minerals to be detected, which are not limited in this embodiment.

[0065] Exemplarily, the energy spectrum acquisition subsystem is used to emit X-rays to the minerals to be detected 13 and detect the X-ray fluorescence and scattered light generated by the minerals to be detected 13 to obtain the initial energy spectrum signal of the minerals to be detected 13 (i.e., the energy spectrum signal of X-ray fluorescence). Refer to Figure 4 As shown, the energy spectrum acquisition subsystem may include, but is not limited to, an X-ray tube 4, a high-voltage controller 5, a dust-proof component 6, and an energy spectrum detector 7. For example, the dust-proof component 6 is used to protect the X-ray tube 4 to prevent dust from entering the X-ray tube 4 and avoid dust contamination of the X-ray tube 4. For example, the dust-proof component 6 can use a mylar film or other types of dust-proof components, as long as it has a dust-proof function. Subsequently, the case where the dust-proof component 6 uses a mylar film is taken as an example for description.

[0066] The high-voltage controller 5 is used to control the X-ray tube 4 to emit X-rays towards the mineral to be detected 13. The X-rays pass through the dust-proof component 6 and reach the surface of the mineral to be detected 13. After the X-rays reach the surface of the mineral to be detected 13, the X-rays interact with the substance of the mineral to be detected 13 to produce the photoelectric effect, Compton scattering, and Rayleigh scattering. In this way, the mineral to be detected 13 can generate X-ray fluorescence and scattered light. For example, the X-ray fluorescence is generated by the photoelectric effect of the X-rays and the substance of the mineral to be detected 13, and the scattered light is generated by Compton scattering and Rayleigh scattering of the X-rays and the substance of the mineral to be detected 13. This process will not be elaborated further.

[0067] The energy spectrum detector 7 is used to detect the X-ray fluorescence and scattered light generated after the mineral to be detected 13 is irradiated by X-rays, and obtain the initial energy spectrum signal of the mineral to be detected 13 (the energy spectrum can also be called the X-ray spectrum). That is, the initial energy spectrum signal can include the energy spectrum information of the X-ray fluorescence and the energy spectrum information of the scattered light.

[0068] For example, the energy spectrum acquisition subsystem can include an X-ray illumination module and an X-ray tube detection module. The X-ray illumination module mainly includes a high-voltage power supply and an X-ray tube (i.e., the high-voltage controller 5 and the X-ray tube 4), etc., and is used to emit X-rays. The X-ray tube detection module mainly includes an X-ray detector (i.e., the energy spectrum detector 7) and a multi-channel analyzer (the multi-channel analyzer is not limited in this embodiment), etc., and is used to efficiently obtain the secondary X-ray fluorescence spectrum generated by the mineral to be detected 13 excited by X-rays, that is, the initial energy spectrum signal.

[0069] For example, when the energy spectrum detector 7 obtains a trigger signal that the mineral to be detected reaches the detection position (i.e., a pre-configured position, when the mineral to be detected reaches the detection position, the acquisition of the initial energy spectrum signal of the mineral to be detected is triggered), it can collect the spectral data of the mineral to be detected and obtain the initial energy spectrum signal (that is, the initial energy spectrum signal is the currently collected spectral data). This initial energy spectrum signal can include the X-ray fluorescence signal and the scattered light signal generated after the mineral to be detected is irradiated by X-rays.

[0070] Exemplarily, referring to Figure 4 As shown, the additional subsystem can include but is not limited to a temperature control unit 12, a radiation shielding protection device 15, and a fan 16. The temperature control unit 12 is used to control the cavity temperature within a certain range (such as less than or equal to a certain temperature threshold), that is, to control the overall temperature of the mineral composition on-line detection system within a certain range to avoid too high internal temperature. The radiation shielding protection device 15 is used for radiation protection. For example, the radiation shielding protection device 15 is connected to the energy spectrum detector 7, and the radiation shielding protection device 15 can prevent the X-rays received by the energy spectrum detector 7 from causing radiation to the outside, that is, the radiation shielding protection device 15 blocks the X-rays.

[0071] The fan 16 is used to blow air on the surface of the mylar film (i.e., the dust-proof component 6) to prevent condensation on the surface of the mylar film. That is, the fan 16 blows the surface of the mylar film to avoid condensation on the surface of the mylar film, reduce the impact of water dew on the energy spectrum signal, and avoid the influence of condensation on the surface of the mylar film on the energy spectrum signal of the X-ray fluorescence signal.

[0072] Exemplarily, the information acquisition subsystem is used to acquire the target correction information corresponding to the initial energy spectrum signal. For example, when the energy spectrum detector 7 obtains the initial energy spectrum signal of the mineral 13 to be detected, it triggers the information acquisition subsystem to acquire the target correction information. Or, when the high-voltage controller 5 controls the X-ray tube 4 to emit X-rays to the mineral 13 to be detected, it triggers the information acquisition subsystem to acquire the target correction information. The information acquisition subsystem includes but is not limited to the laser ranging device 11 (which can be a laser rangefinder or other types of ranging devices, such as a linear array laser rangefinder, a binocular vision area array rangefinder, etc., and there is no limitation on this).

[0073] For example, if it is necessary to correct the influence of the sample preparation surface height on the X-ray fluorescence signal, and / or, if it is necessary to correct the influence of the sample preparation surface roughness on the X-ray fluorescence signal, then the laser ranging device 11 is deployed. If it is not necessary to correct the influence of the sample preparation surface height on the X-ray fluorescence signal, and it is not necessary to correct the influence of the sample preparation surface roughness on the X-ray fluorescence signal, then the laser ranging device 11 is not deployed. Taking the deployment of the laser ranging device 11 as an example.

[0074] The laser ranging device 11 is used to collect the heights of multiple position points on the surface of the mineral 13 to be detected. This height can be the distance between this position point and the laser ranging device 11, or the distance between this position point and the energy spectrum detector 7, or the distance between this position point and the X-ray tube 4, and there is no limitation on this.

[0075] For example, if it is necessary to correct the influence of the ambient temperature on the X-ray fluorescence signal, then the temperature sensor 8 is deployed. If it is not necessary to correct the influence of the ambient temperature on the X-ray fluorescence signal, then the temperature sensor 8 is not deployed. Taking the deployment of the temperature sensor 8 as an example. The temperature sensor 8 is used to collect the current temperature information on the surface of the X-ray tube 4 of the energy spectrum acquisition subsystem. That is, the temperature on the surface of the X-ray tube 4 can be used as the current temperature information.

[0076] For example, if it is necessary to correct the influence of humidity on the X-ray fluorescence signal, then the humidity sensor 14 is deployed. If it is not necessary to correct the influence of humidity on the X-ray fluorescence signal, then the humidity sensor 14 is not deployed. Taking the deployment of the humidity sensor 14 as an example. The humidity sensor 14 is used to collect the current humidity information in the X-ray path (the X-ray path where the X-ray tube 4 emits X-rays) of the energy spectrum acquisition subsystem, that is, the humidity in the X-ray path is used as the current humidity information.

[0077] Exemplarily, the data processing subsystem is used to obtain a target correction coefficient based on the initial energy spectrum signal and the target correction information, and correct the initial energy spectrum signal based on the target correction coefficient to obtain the corrected target energy spectrum signal. The data processing subsystem includes, but is not limited to, a signal control unit 9 and a signal processing unit 10.

[0078] The signal control unit 9 is used to obtain the initial energy spectrum signal from the energy spectrum detector 7, obtain the collected data (such as the heights of multiple position points on the surface of the mineral to be detected 13, the current temperature information, and the current humidity information) from the information acquisition subsystem, determine the target correction information based on the collected data, and transmit the initial energy spectrum signal and the target correction information to the signal processing unit 10. For example, the signal control unit 9 obtains the heights of multiple position points on the surface of the mineral to be detected 13 from the laser ranging device 11, calculates the standard deviation of the heights of the multiple position points, and determines the target roughness of the surface of the mineral to be detected based on the standard deviation, such as the target roughness being the standard deviation. And / or, the signal control unit 9 calculates the average value of the heights of the multiple position points, and determines the target height of the surface of the mineral to be detected based on the average value, such as the target height being the average value. The signal control unit 9 transmits the target roughness and the target height to the signal processing unit 10. The signal control unit 9 obtains the current temperature information from the temperature sensor 8, determines the target temperature based on the current temperature information (such as using the current temperature information as the target temperature, or adjusting and optimizing the current temperature information to obtain the target temperature), and transmits the target temperature to the signal processing unit 10. The signal control unit 9 obtains the current humidity information from the humidity sensor 14, determines the target humidity based on the current humidity information (such as using the current humidity information as the target humidity, or adjusting and optimizing the current humidity information to obtain the target humidity), and transmits the target humidity to the signal processing unit 10.

[0079] The signal processing unit 10 is used to correct the initial energy spectrum signal based on the target correction information to obtain the corrected target energy spectrum signal. For example, the target energy spectrum signal can be used to analyze the material composition of the mineral to be detected. For example, the target correction information can include, but is not limited to, one, any two, any three, or all of the target roughness, the target height, the target temperature, and the target humidity.

[0080] For example, the target roughness and the target height can be referred to as target surface topography correction information (which can also be referred to as mineral surface topography information), and the target surface topography correction information is used to characterize the roughness and / or flatness of the surface of the mineral to be detected, that is, the target surface topography correction information can include the target roughness and / or the target height. When correcting the initial energy spectrum signal based on the target surface topography correction information, dual information characterization of the mineral surface topography and the mineral energy spectrum can be achieved, eliminating the influence of the difference in the surface topography information of the mineral sample on the detection result, achieving high-precision acquisition of the energy spectrum information of the mineral sample, and improving the accuracy of later detection.

[0081] If it is necessary to correct the influence of the height of the sample preparation surface on the X-ray fluorescence signal, the signal processing unit 10 can obtain the target height of the surface of the mineral to be detected, and perform height correction on the initial energy spectrum signal based on the target height. For example, after obtaining the target height of the surface of the mineral to be detected, the signal processing unit 10 queries the height correction information table based on the initial energy spectrum signal and the target height to obtain the target height correction coefficient.

[0082] For example, the height correction information table may include the correspondence between the energy spectrum signal, the height, and the height correction coefficient. As shown in Table 1, it is an example of the height correction information table, and the height correction information table can be represented by the following relationship: aHeight(Hei, i), where Hei is the height of the surface of the mineral to be detected, i is the energy spectrum signal (i.e., the energy magnitude), and aHeight is the correction coefficient of different energy spectrum signals at different heights (denoted as the height correction coefficient). Regarding the acquisition process of the height correction information table, please refer to the subsequent description.

[0083] Table 1

[0084] Energy spectrum signal Height Height correction coefficient S1 H1 CH1 S2 H2 CH2 ... ... ...

[0085] Obviously, after obtaining the initial energy spectrum signal and the target height, by querying the height correction information table shown in Table 1 based on the initial energy spectrum signal and the target height, the target height correction coefficient can be obtained.

[0086] If it is necessary to correct the influence of the surface roughness of the sample preparation on the X-ray fluorescence signal, the signal processing unit 10 can obtain the target roughness of the surface of the mineral to be detected, and perform roughness correction on the initial energy spectrum signal based on the target roughness. For example, after obtaining the target roughness, the signal processing unit 10 queries the roughness correction information table based on the initial energy spectrum signal and the target roughness to obtain the target roughness correction coefficient.

[0087] For example, the roughness correction information table may include the correspondence between the energy spectrum signal, the roughness, and the roughness correction coefficient. As shown in Table 2, it is an example of the roughness correction information table, and the roughness correction information table can be represented by the following relationship: aroughness(Rou, i), where Rou is the roughness of the surface of the mineral to be detected, i is the energy spectrum signal (i.e., the energy magnitude, also known as the energy spectrum), and aroughness is the correction coefficient of different energy spectrum signals at different roughnesses (also can be called the intensity coefficient, denoted as the roughness correction coefficient hereafter). Regarding the acquisition process of the roughness correction information table, please refer to the subsequent description.

[0088] Table 2

[0089] Energy spectrum signal Roughness Roughness correction coefficient S1 R1 CR1 S2 R2 CR2 ... ... ...

[0090] Obviously, after obtaining the initial energy spectrum signal and the target roughness, by querying the roughness correction information table shown in Table 2 based on the initial energy spectrum signal and the target roughness, the target roughness correction coefficient can be obtained.

[0091] If it is necessary to correct the influence of the ambient temperature on the X-ray fluorescence signal, the signal processing unit 10 can obtain the target temperature on the surface of the X-ray tube 4 of the energy spectrum acquisition subsystem, and perform temperature correction on the initial energy spectrum signal based on the target temperature. For example, after the signal processing unit 10 obtains the target temperature on the surface of the X-ray tube 4, it queries the temperature correction information table based on the initial energy spectrum signal and the target temperature to obtain the target temperature correction coefficient.

[0092] For example, the temperature correction information table can include the correspondence between the energy spectrum signal, temperature, and temperature correction coefficient. As shown in Table 3, it is an example of the temperature correction information table, and the temperature correction information table can be expressed by the following relationship: aTemperature(Tem, i), where Tem is the ambient temperature (the temperature on the surface of the X-ray tube 4), i is the energy spectrum signal, and aTemperature is the correction coefficient of different energy spectrum signals at different temperatures (denoted as the temperature correction coefficient). The acquisition process of the temperature correction information table can be referred to the subsequent description.

[0093] Table 3

[0094] Energy spectrum signal Temperature Temperature correction coefficient S1 T1 CT1 S2 T2 CT2 ... ... ...

[0095] Obviously, after obtaining the initial energy spectrum signal and the target temperature, by querying the temperature correction information table shown in Table 3 based on the initial energy spectrum signal and the target temperature, the target temperature correction coefficient can be obtained. On this basis, the working temperature of the X-ray tube 4 can be used to perform temperature correction on the energy spectrum data (initial energy spectrum signal).

[0096] If it is necessary to correct the influence of humidity on the X-ray fluorescence signal, the signal processing unit 10 obtains the target humidity in the X-ray path of the energy spectrum acquisition subsystem, and performs humidity correction on the initial energy spectrum signal based on the target humidity. For example, after the signal processing unit 10 obtains the target humidity in the X-ray path of the energy spectrum acquisition subsystem, it queries the humidity correction information table based on the initial energy spectrum signal and the target humidity to obtain the target humidity correction coefficient.

[0097] For example, the humidity correction information table may include the correspondence between the energy spectrum signal, humidity, and the humidity correction coefficient. As shown in Table 4, which is an example of the humidity correction information table, the humidity correction information table can be represented by the following relationship: aHumidity(Hum, i), where Hum is the ambient relative humidity (humidity in the X-ray path), i is the energy spectrum signal, and aHumidity is the correction coefficient for different energy spectrum signals at different humidities (denoted as the humidity correction coefficient). The process of obtaining the humidity correction information table can be referred to in the following description.

[0098] Table 4

[0099] Energy spectrum signal Humidity Humidity correction coefficient S1 U1 CU1 S2 U2 CU2 ... ... ...

[0100] Obviously, after obtaining the initial energy spectrum signal and the target humidity, by querying the humidity correction information table shown in Table 4 based on the initial energy spectrum signal and the target humidity, the target humidity correction coefficient can be obtained. On this basis, the humidity in the X-ray path can be used to perform humidity correction on the energy spectrum data (initial energy spectrum signal).

[0101] When the signal processing unit 10 corrects the initial energy spectrum signal based on the target correction coefficient to obtain the corrected target energy spectrum signal, the target correction information may include, but is not limited to: one of the target roughness, target height, target temperature, and target humidity, any two, any three, or all of them.

[0102] For example, the target correction coefficient may include one of the target height correction coefficient, target roughness correction coefficient, target temperature correction coefficient, and target humidity correction coefficient. Taking the target roughness correction coefficient as an example, the initial energy spectrum signal can be corrected based on the target roughness correction coefficient to obtain the target energy spectrum signal. For example, the target energy spectrum signal can be the product value between the target roughness correction coefficient and the initial energy spectrum signal.

[0103] For example, the target correction coefficient may simultaneously include two of the target height correction coefficient, target roughness correction coefficient, target temperature correction coefficient, and target humidity correction coefficient. Taking the target roughness correction coefficient and the target temperature correction coefficient as an example, the initial energy spectrum signal can be corrected based on the target roughness correction coefficient and the target temperature correction coefficient to obtain the target energy spectrum signal. For example, the target energy spectrum signal can be the product value among the target roughness correction coefficient, target temperature correction coefficient, and the initial energy spectrum signal.

[0104] For example, the target correction coefficient can include three of the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient. Taking the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient as examples, the initial energy spectrum signal can be corrected based on the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient to obtain the target energy spectrum signal. For example, the target energy spectrum signal can be the product value between the target roughness correction coefficient, the target temperature correction coefficient, the target humidity correction coefficient, and the initial energy spectrum signal.

[0105] For example, the target correction coefficient can include the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient at the same time. The initial energy spectrum signal can be corrected based on the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient to obtain the target energy spectrum signal. For example, the target energy spectrum signal can be the product value between the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient, the target humidity correction coefficient, and the initial energy spectrum signal. For example, the target energy spectrum signal Eccorrection (also known as the corrected energy spectrum signal) is calculated through the following expression: Eccorrection = Ec * aTemperature * aHumidity * aHeight * aroughness. Ec is the initial energy spectrum signal, aTemperature is the target temperature correction coefficient, aHumidity is the target humidity correction coefficient, aHeight is the target height correction coefficient, and aroughness is the target roughness correction coefficient.

[0106] Exemplarily, if the target correction coefficient includes the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient at the same time, based on the relational expression satisfied by several coefficients, the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient can be operated to obtain the comprehensive correction coefficient; after obtaining the comprehensive correction coefficient, the initial energy spectrum signal can be corrected based on the comprehensive correction coefficient to obtain the target energy spectrum signal. For example, the product value of the comprehensive correction coefficient and the initial energy spectrum signal is used as the target energy spectrum signal.

[0107] Regarding the relationship satisfied by several coefficients, this relationship can be pre-configured without any restrictions. For example, the relationship satisfied by several coefficients can be the product value of several coefficients, that is, the product value of the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient is used as the comprehensive correction coefficient. Another example is that the relationship satisfied by several coefficients can be the result of weighted operations on several coefficients, that is, weighted operations are performed on the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient, and the result of the weighted operation is used as the comprehensive correction coefficient. Of course, the above are only examples of the relationship satisfied by several coefficients and can be configured according to requirements.

[0108] If the target correction information includes the target roughness, the target correction coefficient includes the target roughness correction coefficient, and the target roughness correction coefficient is used to characterize the influence degree of the roughness of the surface of the mineral to be detected on the energy spectrum signal. And / or, if the target correction information includes the target height, the target correction coefficient includes the target height correction coefficient, and the target height correction coefficient is used to characterize the influence degree of the height of the surface of the mineral to be detected on the energy spectrum signal. And / or, if the target correction information includes the target temperature, the target correction coefficient includes the target temperature correction coefficient, and the target temperature correction coefficient is used to characterize the influence degree of the surface temperature of the X-ray tube on the energy spectrum signal. And / or, if the target correction information includes the target humidity, the target correction coefficient includes the target humidity correction coefficient, and the target humidity correction coefficient is used to characterize the influence degree of the humidity in the X-ray path of the X-ray tube on the energy spectrum signal.

[0109] In the above embodiments, a roughness correction information table, a temperature correction information table, a humidity correction information table, and a height correction information table are involved. See Tables 1-4. The acquisition processes of the roughness correction information table, the temperature correction information table, the humidity correction information table, and the height correction information table are described below.

[0110] Place the sample object on the conveying device 1. The sample object can be coal, etc. To distinguish it from the mineral to be detected, the substance to be measured for obtaining each correction information table is called the sample object. When the sample object is conveyed by the conveying device 1 to the scraper 2, the scraper 2 performs a leveling operation on the sample object. When the sample object is conveyed by the conveying device 1 to the pressure roller 3, the pressure roller 3 performs a flattening operation on the sample object.

[0111] The high-voltage controller 5 controls the X-ray tube 4 to emit X-rays towards the sample object. The X-rays pass through the dust-proof component and reach the surface of the sample object. The sample object can generate X-ray fluorescence and scattered light. The energy spectrum detector 7 detects the X-ray fluorescence and scattered light generated by the sample object to obtain the energy spectrum signal of the sample object, and this energy spectrum signal can include the energy spectrum information of the X-ray fluorescence and the energy spectrum information of the scattered light.

[0112] The laser distance measuring device 11 collects the heights of multiple position points on the surface of the sample object, calculates the standard deviation of the heights of the multiple position points, and determines the surface roughness of the sample object based on this standard deviation (such as using this standard deviation as the surface roughness of the sample object). The laser distance measuring device 11 calculates the average value of the heights of the multiple position points and determines the surface height of the sample object based on this average value (such as using this average value as the surface height of the sample object).

[0113] The temperature sensor 8 collects the temperature on the surface of the X-ray tube 4 of the energy spectrum acquisition subsystem.

[0114] The humidity sensor 14 collects the humidity in the X-ray path of the energy spectrum acquisition subsystem.

[0115] The signal processing unit 10 can collect the above information and construct sample data of the sample object based on the above information. The sample data can include the energy spectrum signal of the sample object obtained by the energy spectrum detector 7, the surface roughness and surface height of the sample object obtained by the laser distance measuring device 11, the temperature on the surface of the X-ray tube 4 obtained by the temperature sensor 8, and the humidity in the X-ray path obtained by the humidity sensor 14.

[0116] Exemplarily, the high-voltage controller 5 can control the X-ray tube 4 to emit multiple X-rays with different doses to the sample object (the dose represents the number of photons of the emitted X-rays. If the dose is larger, the number of photons of the emitted X-rays is more; if the dose is smaller, the number of photons of the emitted X-rays is less).

[0117] For each X-ray, a sample data can be constructed in the above manner, that is, multiple X-rays correspond to multiple sample data. In this way, multiple sample data of the sample object can be obtained. For each sample data, the sample data can include the energy spectrum signal of the sample object, the surface roughness of the sample object, the surface height of the sample object, the temperature on the surface of the X-ray tube 4, and the humidity in the X-ray path.

[0118] One sample data can be selected from the multiple sample data as the standard sample data, that is, the standard sample data can be any one of the multiple sample data, and there is no restriction on this standard sample data.

[0119] For each sample data, the quotient between the roughness in this sample data and the roughness in the standard sample data can be calculated, and this quotient is used as the roughness correction coefficient, and the roughness correction coefficient is recorded in this sample data, indicating the roughness correction coefficient corresponding to the roughness and energy spectrum signal in this sample data.

[0120] For each sample data, the quotient value between the temperature in the sample data and the temperature in the standard sample data can be calculated. This quotient value is used as the temperature correction coefficient, and this temperature correction coefficient is recorded in the sample data, indicating the temperature in the sample data and the temperature correction coefficient corresponding to the energy spectrum signal.

[0121] For each sample data, the quotient value between the humidity in the sample data and the humidity in the standard sample data can be calculated. This quotient value is used as the humidity correction coefficient, and this humidity correction coefficient is recorded in the sample data, indicating the humidity in the sample data and the humidity correction coefficient corresponding to the energy spectrum signal.

[0122] For each sample data, the quotient value between the height in the sample data and the height in the standard sample data can be calculated. This quotient value is used as the height correction coefficient, and this height correction coefficient is recorded in the sample data, indicating the height in the sample data and the height correction coefficient corresponding to the energy spectrum signal.

[0123] In summary, for each sample data, the sample data may include but is not limited to the following: the energy spectrum signal of the sample object, the surface roughness of the sample object, the roughness correction coefficient corresponding to the roughness and the energy spectrum signal, the height of the surface of the sample object, the height correction coefficient corresponding to the height and the energy spectrum signal, the temperature on the surface of the X-ray tube 4, the temperature correction coefficient corresponding to the height and the energy spectrum signal, the humidity in the X-ray path, and the humidity correction coefficient corresponding to the height and the energy spectrum signal.

[0124] On this basis, for each sample data, the corresponding relationship between the energy spectrum signal in the sample data, the roughness in the sample data, and the roughness correction coefficient in the sample data can be recorded in the roughness correction information table, as shown in Table 2. The corresponding relationship between the energy spectrum signal in the sample data, the temperature in the sample data, and the temperature correction coefficient in the sample data can be recorded in the temperature correction information table, as shown in Table 3. The corresponding relationship between the energy spectrum signal in the sample data, the humidity in the sample data, and the humidity correction coefficient in the sample data can be recorded in the humidity correction information table, as shown in Table 4. The corresponding relationship between the energy spectrum signal in the sample data, the height in the sample data, and the height correction coefficient in the sample data can be recorded in the height correction information table, as shown in Table 1.

[0125] In a possible implementation, after the signal processing unit 10 corrects the initial energy spectrum signal based on the target correction coefficient to obtain the target energy spectrum signal, the target energy spectrum signal is used to analyze the material composition of the mineral to be detected. For example, the target energy spectrum signal is input into a trained neural network model (such as a bimodal neural network inference model) so that the neural network model is based on the signal characteristics of the target energy spectrum signal, and the signal characteristics are used to analyze the material composition of the mineral to be detected. For example, the signal characteristics represent the characteristics of the material composition, and the material composition of the mineral to be detected can be directly determined based on the signal characteristics. Or, indirect calculation can also be performed based on the signal characteristics, and the material composition of the mineral to be detected is determined based on the indirect calculation result. There is no limitation on this, as long as the material composition of the mineral to be detected can be analyzed based on the signal characteristics.

[0126] For example, the material composition may include, but is not limited to, one or more of ash component, ash content, volatile matter, carbon and hydrogen, ash fusion temperature, total water, total sulfur, and calorific value.

[0127] Exemplarily, for the training process of the neural network, the neural network can be trained in the following way: Based on the known true material composition of the substance to be measured, the sample energy spectrum signal corresponding to the substance to be measured and the sample label corresponding to the substance to be measured can be obtained. The acquisition method of the sample energy spectrum signal is the same as that of the target energy spectrum signal. Performing the above processing on the substance to be measured can obtain the sample energy spectrum signal, which will not be repeated here. The sample label can represent the true material composition corresponding to the substance to be measured.

[0128] The sample energy spectrum signal (such as multiple sample energy spectrum signals) can be input into the neural network to be trained, and the neural network determines the predicted material composition corresponding to the substance to be measured based on the sample energy spectrum signal. There is no limitation on this process. Then, the loss value can be calculated based on the true material composition corresponding to the substance to be measured and the predicted material composition corresponding to the substance to be measured. The greater the difference between the true material composition and the predicted material composition, the greater the loss value; the smaller the difference between the true material composition and the predicted material composition, the smaller the loss value.

[0129] The network parameters (network weights) of the neural network can be adjusted based on the loss value. The adjustment goal is to make the loss value smaller and smaller until the neural network converges to obtain the trained neural network.

[0130] Based on the trained neural network, after the signal processing unit 10 obtains the target energy spectrum signal, the target energy spectrum signal is input into the trained neural network. The neural network determines the material composition of the mineral to be detected based on the target energy spectrum signal and outputs the material composition of the mineral to be detected. There is no limitation on this process.

[0131] Exemplarily, the on-line mineral composition detection system further includes a spectral subsystem (such as a near-infrared spectrometer). The spectral subsystem is used to collect the near-infrared spectral signals of the mineral to be detected. The signal control unit is used to obtain the near-infrared spectral signals from the spectral subsystem and send the near-infrared spectral signals to the signal processing unit. The signal processing unit is further used to, after obtaining the target energy spectrum signal and the near-infrared spectral signal, input the target energy spectrum signal and the near-infrared spectral signal into the trained neural network model, so that the neural network model outputs signal features based on the target energy spectrum signal and the near-infrared spectral signal. For example, the neural network model fuses the target energy spectrum signal and the near-infrared spectral signal to obtain a fused signal, and determines and outputs signal features based on the fused signal, and the signal features are used to analyze the material composition of the mineral to be detected.

[0132] For example, the fused signal is the fusion of the near-infrared spectral signal and the target energy spectrum signal in a dual-spectrum manner. The fused signal can also be the fusion of three spectra, such as the combination of near-infrared spectral signal + target energy spectrum signal + a third one, such as the combination of microwave + near-infrared spectral signal + target energy spectrum signal, the combination of dual X-ray fluorescence (i.e., two frames of energy spectrum signals) + near-infrared spectral signal, etc.

[0133] On this basis, the material composition can be determined based on the fused signal. For example, the near-infrared spectral signal can be used to sense the signals of molecular groups related to carbon, hydrogen, and oxygen elements in the mineral to be detected (such as coal), and relevant information such as fixed carbon, volatile matter, and moisture in the industrial composition of coal can be correlated. Using the XRF energy spectrum (i.e., the target energy spectrum signal), information on various elements (Al, Si, Ca, Fe, etc.) related to ash and sulfur content information in coal can be obtained. And the near-infrared spectrum and energy spectrum data are fused with each other.

[0134] The fusion methods can include: 1. First, based on the XRF energy spectrum data combined with deep learning, calculate the ash value, then splice the ash value into the near-infrared spectral data, and use the deep learning algorithm to automatically learn the weights of the spectral data and the ash data, and perform information fusion of the two dimensions to output the calorific value. 2. Directly splice the preprocessed energy spectrum data and spectral data, combine the deep learning algorithm, automatically learn the weights of the spectral data and the energy spectrum data, and perform information fusion of the two dimensions to output the calorific value.

[0135] The molecular information in the mineral to be detected (such as coal) can be detected by near-infrared light to generate a molecular spectrum, and the atomic information in the mineral to be detected can be detected by X-ray fluorescence to generate an atomic spectrum. Further, various components in the mineral to be detected can be detected based on the molecular spectrum and the atomic spectrum. The various components here include carbon, hydrogen, nitrogen, sulfur, moisture, ash, etc., and parameter information such as calorific value can also be obtained. Optionally, according to the atomic spectrum and the molecular spectrum, a neural network model technology is used to detect various components. Specifically, near-infrared spectroscopy technology and X-ray fluorescence spectroscopy technology are combined, and an artificial intelligence algorithm is used to effectively fuse the molecular spectrum and the atomic spectrum, thereby realizing rapid, accurate, and non-destructive analysis of mineral components.

[0136] As can be seen from the above technical solutions, in the embodiment of the present application, after obtaining the initial energy spectrum signal of X-ray fluorescence, the initial energy spectrum signal is corrected based on the target correction coefficient to obtain the target energy spectrum signal. The interference information in the initial energy spectrum signal is excluded through correction, and then the material composition of the mineral to be detected is analyzed based on the target energy spectrum signal, and an accurate material composition result can be obtained, avoiding obtaining an incorrect material composition result. For example, roughness interference information, height interference information, temperature interference information, and humidity interference information in the initial energy spectrum signal can be excluded, so that an accurate material composition result can be obtained. The temperature correction of the energy spectrum data can be performed using the working temperature data of the X-ray tube to eliminate the influence of different tube temperatures on the energy spectrum signal. The humidity correction of the energy spectrum data can be performed using the humidity in the X-ray path to eliminate the influence of humidity on the energy spectrum signal. The influence of sample preparation differences can be eliminated using the surface topography information (roughness and height) of the substance to be measured.

[0137] Based on the same application concept as the above system, a method for on-line detection of mineral components is proposed in the embodiment of the present application. Refer to Figure 5 As shown, it is a schematic flow chart of the method. The method may include:

[0138] Step 501: Emit X-rays to the mineral to be detected, and detect the X-ray fluorescence and scattered light generated after the mineral to be detected is irradiated by the X-rays to obtain the initial energy spectrum signal of the mineral to be detected.

[0139] Step 502: Obtain the surface topography information of the mineral to be detected, and the topography information is used to characterize the roughness and / or flatness of the surface of the mineral to be detected.

[0140] Step 503: Correct the initial energy spectrum signal based on the topography information to obtain the corrected target energy spectrum signal, and the target energy spectrum signal is used to analyze the material composition of the mineral to be detected.

[0141] Exemplarily, the topography information includes the heights of multiple position points on the surface of the mineral to be detected; correcting the initial energy spectrum signal based on the topography information to obtain the corrected target energy spectrum signal may include: determining the target roughness based on the variance of the heights of the multiple position points; querying the obtained roughness correction information table based on the initial energy spectrum signal and the target roughness to obtain the target roughness correction coefficient, where the target roughness correction coefficient is used to characterize the influence degree of the roughness of the surface of the mineral to be detected on the energy spectrum signal; wherein, the roughness correction information table includes the corresponding relationship between the energy spectrum signal, the roughness, and the roughness correction coefficient. Correcting the initial energy spectrum signal based on the target roughness correction coefficient to obtain the corrected target energy spectrum signal.

[0142] Exemplarily, the process of obtaining the roughness correction information table may include, but is not limited to: obtaining multiple sample data of a sample object, and for each sample data, the sample data includes a roughness correction coefficient, an energy spectrum signal, and a roughness; wherein, the roughness correction coefficient is the quotient of the roughness in the sample data and the roughness in the standard sample data, and the standard sample data is any one of the multiple sample data; recording the corresponding relationship between the energy spectrum signal, the roughness, and the roughness correction coefficient in the sample data in the roughness correction information table.

[0143] Exemplarily, it may further include: determining the target height based on the average value of the heights of the multiple position points; querying the obtained height correction information table based on the initial energy spectrum signal and the target height to obtain the target height correction coefficient, where the target height correction coefficient is used to characterize the influence degree of the height of the surface of the mineral to be detected on the energy spectrum signal; wherein, the height correction information table includes the corresponding relationship between the energy spectrum signal, the height, and the height correction coefficient; correcting the initial energy spectrum signal based on the target roughness correction coefficient and the target height correction coefficient to obtain the corrected target energy spectrum signal.

[0144] For example, the process of obtaining the height correction information table may include, but is not limited to: obtaining multiple sample data of a sample object, and for each sample data, the sample data includes a height correction coefficient, the obtained energy spectrum signal, and a height; the height correction coefficient is the quotient of the height in the sample data and the height in the standard sample data, and the standard sample data is any one of the multiple sample data; recording the corresponding relationship between the energy spectrum signal, the height, and the height correction coefficient in the sample data in the height correction information table.

[0145] Exemplarily, it may further include: obtaining current temperature information, determining a target temperature based on the current temperature information; querying an acquired temperature correction information table based on the initial energy spectrum signal and the target temperature to obtain a target temperature correction coefficient, where the target temperature correction coefficient is used to characterize the influence degree of the surface temperature on the energy spectrum signal; wherein, the temperature correction information table includes the corresponding relationship between the energy spectrum signal, temperature, and temperature correction coefficient; correcting the initial energy spectrum signal based on the target roughness correction coefficient and the target temperature correction coefficient to obtain a corrected target energy spectrum signal.

[0146] For example, the process of obtaining the temperature correction information table may include, but is not limited to: obtaining multiple sample data of a sample object, and for each sample data, the sample data includes a temperature correction coefficient, an energy spectrum signal, and a temperature; the temperature correction coefficient is the quotient between the temperature in the sample data and the temperature in the standard sample data, and the standard sample data is any one of the multiple sample data; recording the corresponding relationship between the energy spectrum signal, the temperature, and the temperature correction coefficient in the sample data in the temperature correction information table.

[0147] Exemplarily, it may further include: obtaining current humidity information, determining a target humidity based on the current humidity information; querying an acquired humidity correction information table based on the initial energy spectrum signal and the target humidity to obtain a target humidity correction coefficient, where the target humidity correction coefficient is used to characterize the influence degree of the humidity on the energy spectrum signal; wherein, the humidity correction information table includes the corresponding relationship between the energy spectrum signal, humidity, and humidity correction coefficient; correcting the initial energy spectrum signal based on the target roughness correction coefficient and the target humidity correction coefficient to obtain a corrected target energy spectrum signal.

[0148] For example, the process of obtaining the humidity correction information table may include, but is not limited to: obtaining multiple sample data of a sample object, and for each sample data, the sample data includes a humidity correction coefficient, an energy spectrum signal, and a humidity; the humidity correction coefficient is the quotient between the humidity in the sample data and the humidity in the standard sample data, and the standard sample data is any one of the multiple sample data; recording the corresponding relationship between the energy spectrum signal, the humidity, and the humidity correction coefficient in the sample data in the humidity correction information table.

[0149] Exemplarily, the target correction coefficient may include one, any two, any three, or all of a target height correction coefficient, a target roughness correction coefficient, a target temperature correction coefficient, and a target humidity correction coefficient. Based on this, if the target correction coefficient includes the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient at the same time, then based on the relationship satisfied by several coefficients, operations are performed on the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient, and the target humidity correction coefficient (such as calculating the product value of these coefficients) to obtain a comprehensive correction coefficient; the initial energy spectrum signal is corrected based on the comprehensive correction coefficient to obtain the target energy spectrum signal.

[0150] Based on the same inventive concept as the above system, an on-line mineral composition detection device is proposed in an embodiment of the present application. Refer to Figure 6 As shown in the figure, which is a schematic structural diagram of the device, the device may include:

[0151] An acquisition module 61, configured to acquire an initial energy spectrum signal and the topographic information of the surface of the mineral to be detected, where the topographic information is used to characterize the roughness and / or flatness of the surface of the mineral to be detected; wherein, the initial energy spectrum signal is obtained by detecting the X-ray fluorescence and scattered light generated after the mineral to be detected is irradiated by X-rays; a correction module 62, configured to correct the initial energy spectrum signal based on the topographic information to obtain a corrected target energy spectrum signal, and the target energy spectrum signal is used to analyze the material composition of the mineral to be detected.

[0152] Exemplarily, when the correction module 62 corrects the initial energy spectrum signal based on the topographic information to obtain a corrected target energy spectrum signal, it is specifically configured to: determine the target roughness based on the variance of the heights of multiple position points; query the obtained roughness correction information table based on the initial energy spectrum signal and the target roughness to obtain a target roughness correction coefficient, where the target roughness correction coefficient is used to characterize the influence degree of the roughness of the surface of the mineral to be detected on the energy spectrum signal; wherein, the roughness correction information table includes the corresponding relationship between the energy spectrum signal, the roughness, and the roughness correction coefficient. The initial energy spectrum signal is corrected based on the target roughness correction coefficient to obtain a corrected target energy spectrum signal. For example, the acquisition process of the roughness correction information table may include, but is not limited to: acquiring multiple sample data of a sample object, and for each sample data, the sample data includes a roughness correction coefficient, an energy spectrum signal, and a roughness; wherein, the roughness correction coefficient is the quotient value between the roughness in the sample data and the roughness in the standard sample data, and the standard sample data is any one of the multiple sample data; record the corresponding relationship between the energy spectrum signal in the sample data, the roughness in the sample data, and the roughness correction coefficient in the roughness correction information table.

[0153] Exemplarily, the calibration module 62 is further configured to: determine a target height based on the average value of the heights of multiple position points, query the acquired height calibration information table based on the initial energy spectrum signal and the target height, and obtain a target height calibration coefficient, where the target height calibration coefficient is used to characterize the influence degree of the height of the surface of the mineral to be detected on the energy spectrum signal; wherein, the height calibration information table includes the correspondence between the energy spectrum signal, the height and the height calibration coefficient; calibrate the initial energy spectrum signal based on the target roughness calibration coefficient and the target height calibration coefficient to obtain the calibrated target energy spectrum signal. For example, the process of obtaining the height calibration information table may include, but is not limited to: obtaining multiple sample data of a sample object, and for each sample data, the sample data includes a height calibration coefficient, the acquired energy spectrum signal and the height; the height calibration coefficient is the quotient between the height in the sample data and the height in the standard sample data, and the standard sample data is any one of the multiple sample data; record the correspondence between the energy spectrum signal in the sample data, the height in the sample data and the height calibration coefficient in the height calibration information table.

[0154] Exemplarily, the calibration module 62 is further configured to obtain current temperature information, and determine a target temperature based on the current temperature information; query the acquired temperature calibration information table based on the initial energy spectrum signal and the target temperature, and obtain a target temperature calibration coefficient, where the target temperature calibration coefficient is used to characterize the influence degree of the surface temperature on the energy spectrum signal; wherein, the temperature calibration information table includes the correspondence between the energy spectrum signal, the temperature and the temperature calibration coefficient; calibrate the initial energy spectrum signal based on the target roughness calibration coefficient and the target temperature calibration coefficient to obtain the calibrated target energy spectrum signal.

[0155] For example, the process of obtaining the temperature calibration information table may include, but is not limited to: obtaining multiple sample data of a sample object, and for each sample data, the sample data includes a temperature calibration coefficient, the energy spectrum signal and the temperature; the temperature calibration coefficient is the quotient between the temperature in the sample data and the temperature in the standard sample data, and the standard sample data is any one of the multiple sample data; record the correspondence between the energy spectrum signal in the sample data, the temperature in the sample data and the temperature calibration coefficient in the temperature calibration information table.

[0156] Exemplarily, the calibration module 62 is further configured to obtain current humidity information, and determine a target humidity based on the current humidity information; query the acquired humidity calibration information table based on the initial energy spectrum signal and the target humidity, and obtain a target humidity calibration coefficient, where the target humidity calibration coefficient is used to characterize the influence degree of the humidity on the energy spectrum signal; wherein, the humidity calibration information table includes the correspondence between the energy spectrum signal, the humidity and the humidity calibration coefficient; calibrate the initial energy spectrum signal based on the target roughness calibration coefficient and the target humidity calibration coefficient to obtain the calibrated target energy spectrum signal.

[0157] For example, the process of obtaining the humidity correction information table may include, but is not limited to: obtaining a plurality of sample data of a sample object, and for each sample data, the sample data includes a humidity correction coefficient, an energy spectrum signal, and humidity; the humidity correction coefficient is the quotient between the humidity in the sample data and the humidity in the standard sample data, and the standard sample data is any one of the plurality of sample data; recording the corresponding relationship between the energy spectrum signal in the sample data, the humidity in the sample data, and the humidity correction coefficient in the sample data in the humidity correction information table.

[0158] Based on the same application concept as the above method, an electronic device is proposed in an embodiment of the present application, including: a processor and a machine-readable storage medium, the machine-readable storage medium stores machine-executable instructions that can be executed by the processor; the processor is used to execute the machine-executable instructions to implement the online mineral composition detection method disclosed in the above examples of the present application.

[0159] Based on the same application concept as the above method, an embodiment of the present application also provides a machine-readable storage medium, on which a number of computer instructions are stored, and when the computer instructions are executed by a processor, the online mineral composition detection method disclosed in the above examples of the present application can be implemented.

[0160] Among them, the above machine-readable storage medium can be any electronic, magnetic, optical or other physical storage device that can contain or store information, such as executable instructions, data, etc. For example, the machine-readable storage medium can be: RAM (Random Access Memory), volatile memory, non-volatile memory, flash memory, storage drives (such as hard disk drives), solid state drives, any type of storage disk (such as optical discs, DVDs, etc.), or similar storage media, or a combination thereof.

[0161] An embodiment of the present application also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the online mineral composition detection method disclosed in the above examples of the present application can be implemented.

[0162] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. The embodiments of the present application can take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code.

[0163] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, various modifications and variations can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A mineral composition online detection system, characterized in that: The system comprises at least: Sample preparation system, used for online sample preparation of minerals to be tested; The energy spectrum acquisition subsystem is located downstream of the sample preparation system, and the energy spectrum acquisition subsystem at least includes an X-ray tube, a high-voltage controller and an energy spectrum detector; the high-voltage controller is used to control the X-ray tube to emit X-rays to the mineral to be detected; the energy spectrum detector is used to collect spectrum data of the mineral to be detected when a trigger signal indicating that the mineral to be detected has reached the detection position is obtained, so as to obtain an initial energy spectrum signal of the mineral to be detected; wherein the initial energy spectrum signal includes an X-ray fluorescence signal and a scattered light signal generated by the mineral to be detected after being irradiated by the X-rays; An information collection subsystem, the information collection subsystem comprising a laser distance measuring device; the laser distance measuring device is used to collect the morphological information of the surface of the mineral to be detected; A data processing subsystem is used to obtain the initial energy spectrum signal and the morphology information, wherein the morphology information is used to characterize the roughness and / or flatness of the surface of the mineral to be detected; based on the morphology information, the initial energy spectrum signal is corrected to obtain a corrected target energy spectrum signal; Wherein, the target energy spectrum signal is used to analyze the material composition of the mineral to be detected.

2. The system according to claim 1, characterized in that The morphological information includes the heights of multiple position points on the surface of the mineral to be detected; the data processing subsystem corrects the initial energy spectrum signal based on the morphological information to obtain the corrected target energy spectrum signal, which is specifically used for: The target roughness is determined based on the variance of the heights of the multiple position points; based on the initial energy spectrum signal and the target roughness, the obtained roughness correction information table is queried to obtain a target roughness correction coefficient, and the target roughness correction coefficient is used to characterize the influence of the roughness of the surface of the mineral to be detected on the energy spectrum signal; wherein the roughness correction information table includes the corresponding relationship between the energy spectrum signal, the roughness and the roughness correction coefficient; The initial energy spectrum signal is corrected based on the target roughness correction coefficient to obtain the corrected target energy spectrum signal.

3. The system according to claim 2, characterized in that The data processing subsystem corrects the initial energy spectrum signal based on the morphology information to obtain the corrected target energy spectrum signal, which is specifically used for: The target height is determined based on the average value of the heights of the multiple position points; based on the initial energy spectrum signal and the target height, the acquired height correction information table is queried to obtain a target height correction coefficient, and the target height correction coefficient is used to characterize the influence of the height of the surface of the mineral to be detected on the energy spectrum signal; wherein the height correction information table includes the corresponding relationship between the energy spectrum signal, the height and the height correction coefficient; The initial energy spectrum signal is corrected based on the target roughness correction coefficient and the target height correction coefficient to obtain the corrected target energy spectrum signal.

4. The system according to claim 2, characterized in that The information collection subsystem further includes a temperature sensor, and the temperature sensor is used to collect current temperature information of the surface of the X-ray tube; The data processing subsystem is further used to obtain the current temperature information, determine the target temperature based on the current temperature information; query the acquired temperature correction information table based on the initial energy spectrum signal and the target temperature to obtain a target temperature correction coefficient, wherein the target temperature correction coefficient is used to characterize the influence of the surface temperature of the X-ray tube on the energy spectrum signal; wherein the temperature correction information table includes a corresponding relationship between the energy spectrum signal, the temperature and the temperature correction coefficient; The initial energy spectrum signal is corrected based on the target roughness correction coefficient and the target temperature correction coefficient to obtain the corrected target energy spectrum signal.

5. The system according to claim 2, characterized in that If the target correction information also includes target humidity, the information acquisition subsystem further includes a humidity sensor, and the humidity sensor is used to collect current humidity information in the X-ray path; The data processing subsystem is further used to obtain the current humidity information, determine the target humidity based on the current humidity information; query the acquired humidity correction information table based on the initial energy spectrum signal and the target humidity to obtain a target humidity correction coefficient, wherein the target humidity correction coefficient is used to characterize the degree of influence of humidity in the X-ray path of the X-ray tube on the energy spectrum signal; wherein the humidity correction information table includes a corresponding relationship between the energy spectrum signal, humidity and the humidity correction coefficient; The initial energy spectrum signal is corrected based on the target roughness correction coefficient and the target humidity correction coefficient to obtain the corrected target energy spectrum signal.

6. The system according to any one of claims 2 to 5, characterized in that: The data processing subsystem is further used to calculate the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient and the target humidity correction coefficient based on a relationship satisfied by the target height correction coefficient, the target roughness correction coefficient, the target temperature correction coefficient and the target humidity correction coefficient to obtain a comprehensive correction coefficient; The initial energy spectrum signal is corrected based on the comprehensive correction coefficient to obtain a target energy spectrum signal.

7. The system according to claim 1, characterized in that The sample making system comprises a detection belt, a scraper and a pressure roller; The detection belt is used to convey the mineral to be detected; The scraper is used to perform a scraping operation on the mineral to be detected when the detection belt conveys the mineral to be detected to the scraper; the pressing roller is used to perform a pressing operation on the mineral to be detected when the detection belt conveys the mineral to be detected to the pressing roller; Wherein, the online sample preparation operation includes the scraping operation and the flattening operation.

8. The system according to claim 1, characterized in that The system further comprises: an additional subsystem, wherein the additional subsystem comprises a temperature control unit, and the temperature control unit is used to control the temperature of the mineral composition online detection system within a preset temperature range.

9. The system according to claim 1, characterized in that The system further includes: an additional subsystem, wherein the additional subsystem includes a radiation shielding protection device, and the radiation shielding protection device is used to prevent the X-rays received by the energy spectrum detector from causing radiation to the outside.

10. The system according to claim 1, characterized in that The system further comprises: an additional subsystem, the additional subsystem comprising a fan; The energy spectrum acquisition subsystem further includes a dustproof component, which is located between the X-ray tube and the mineral to be detected, and the X-rays pass through the dustproof component to reach the surface of the mineral to be detected; Wherein, if the dust-proof component is a Mylar film, the fan is used to blow air on the surface of the Mylar film to avoid condensation on the surface of the Mylar film.

11. The system according to claim 1, characterized in that The data processing subsystem is further used for, after obtaining the target energy spectrum signal, inputting the target energy spectrum signal into a trained neural network model so that the neural network model outputs signal features based on the target energy spectrum signal; or, inputting the target energy spectrum signal and a near-infrared spectrum signal into the neural network model so that the neural network model outputs signal features based on the target energy spectrum signal and the near-infrared spectrum signal; The signal characteristics are used to analyze the material composition of the mineral to be detected; the material composition includes: one or more of ash composition, ash content, volatile matter, carbon and hydrogen, ash melting point, total water, total sulfur and calorific value; The mineral composition online detection system further includes a spectral subsystem, wherein the spectral subsystem is used to collect near-infrared spectral signals of the mineral to be detected, and the data processing subsystem is used to obtain the near-infrared spectral signals.

12. A method for online detection of mineral components, characterized in that: The method comprises: Emitting X-rays to the mineral to be detected, and detecting X-ray fluorescence and scattered light generated by the mineral to be detected after being irradiated by the X-rays, to obtain an initial energy spectrum signal of the mineral to be detected; Acquiring morphological information of the surface of the mineral to be detected, wherein the morphological information is used to characterize the roughness and / or flatness of the surface of the mineral to be detected; The initial energy spectrum signal is corrected based on the morphological information to obtain a corrected target energy spectrum signal, and the target energy spectrum signal is used to analyze the material composition of the mineral to be detected.

13. The method according to claim 12, characterized in that The topographic information includes the heights of a plurality of position points on the surface of the mineral to be detected; The correcting the initial energy spectrum signal based on the morphology information to obtain a corrected target energy spectrum signal includes: determining a target roughness based on a variance of the heights of the plurality of location points; Based on the initial energy spectrum signal and the target roughness, the obtained roughness correction information table is queried to obtain a target roughness correction coefficient, wherein the target roughness correction coefficient is used to characterize the influence of the roughness of the surface of the mineral to be detected on the energy spectrum signal; wherein the roughness correction information table includes the corresponding relationship between the energy spectrum signal, the roughness and the roughness correction coefficient; Correcting the initial energy spectrum signal based on the target roughness correction coefficient to obtain the corrected target energy spectrum signal; Wherein, the acquisition process of the roughness correction information table includes: acquiring multiple sample data of the sample object, for each sample data, the sample data includes a roughness correction coefficient, an energy spectrum signal and roughness; wherein the roughness correction coefficient is a quotient between the roughness in the sample data and the roughness in the standard sample data, and the standard sample data is any sample data among the multiple sample data; recording the correspondence between the energy spectrum signal in the sample data, the roughness in the sample data and the roughness correction coefficient in the sample data in the roughness correction information table.

14. The method according to claim 13, characterized in that The method further comprises: Determining a target height based on an average of the heights of the plurality of location points; Based on the initial energy spectrum signal and the target height, the acquired height correction information table is queried to obtain a target height correction coefficient, wherein the target height correction coefficient is used to characterize the influence of the height of the mineral surface to be detected on the energy spectrum signal; wherein the height correction information table includes a corresponding relationship between the energy spectrum signal, the height and the height correction coefficient; based on the target roughness correction coefficient and the target height correction coefficient, the initial energy spectrum signal is corrected to obtain the corrected target energy spectrum signal; Among them, the acquisition process of the height correction information table includes: acquiring multiple sample data of the sample object, for each sample data, the sample data includes a height correction coefficient, an acquired energy spectrum signal and a height; the height correction coefficient is a quotient between the height in the sample data and the height in the standard sample data, and the standard sample data is any sample data among the multiple sample data; and recording in the height correction information table the correspondence between the energy spectrum signal in the sample data, the height in the sample data and the height correction coefficient in the sample data.

15. The method according to claim 13, characterized in that The method further comprises: Acquiring current temperature information, and determining a target temperature based on the current temperature information; Based on the initial energy spectrum signal and the target temperature, the acquired temperature correction information table is queried to obtain a target temperature correction coefficient, and the target temperature correction coefficient is used to characterize the influence of the surface temperature on the energy spectrum signal; wherein the temperature correction information table includes the corresponding relationship between the energy spectrum signal, the temperature and the temperature correction coefficient; based on the target roughness correction coefficient and the target temperature correction coefficient, the initial energy spectrum signal is corrected to obtain the corrected target energy spectrum signal; Among them, the process of obtaining the temperature correction information table includes: obtaining multiple sample data of the sample object, for each sample data, the sample data includes a temperature correction coefficient, an energy spectrum signal and a temperature; the temperature correction coefficient is a quotient between the temperature in the sample data and the temperature in the standard sample data, and the standard sample data is any sample data among the multiple sample data; recording the correspondence between the energy spectrum signal in the sample data, the temperature in the sample data and the temperature correction coefficient in the sample data in the temperature correction information table.

16. The method according to claim 13, characterized in that The method further comprises: Acquiring current humidity information, and determining a target humidity based on the current humidity information; Based on the initial energy spectrum signal and the target humidity, the humidity correction information table that has been obtained is queried to obtain a target humidity correction coefficient, and the target humidity correction coefficient is used to characterize the degree of influence of humidity on the energy spectrum signal; wherein the humidity correction information table includes a corresponding relationship between the energy spectrum signal, humidity and the humidity correction coefficient; based on the target roughness correction coefficient and the target humidity correction coefficient, the initial energy spectrum signal is corrected to obtain the corrected target energy spectrum signal; Among them, the acquisition process of the humidity correction information table includes: acquiring multiple sample data of the sample object, for each sample data, the sample data includes a humidity correction coefficient, an energy spectrum signal and humidity; the humidity correction coefficient is the quotient between the humidity in the sample data and the humidity in the standard sample data, and the standard sample data is any sample data among the multiple sample data; recording the correspondence between the energy spectrum signal in the sample data, the humidity in the sample data and the humidity correction coefficient in the sample data in the humidity correction information table.

17. An online detection device for mineral components, characterized in that: The device comprises: An acquisition module, used to acquire an initial energy spectrum signal and morphological information of the surface of the mineral to be detected, wherein the morphological information is used to characterize the roughness and / or flatness of the surface of the mineral to be detected; wherein the initial energy spectrum signal is obtained by detecting X-ray fluorescence and scattered light generated by the mineral to be detected after being irradiated with X-rays; The correction module is used to correct the initial energy spectrum signal based on the morphology information to obtain a corrected target energy spectrum signal, and the target energy spectrum signal is used to analyze the material composition of the mineral to be detected.

18. An electronic device, characterized in that: include: a processor and a machine-readable storage medium storing machine-executable instructions executable by the processor; The processor is used to execute machine executable instructions to implement the method described in any one of claims 12-16.

19. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores machine-executable instructions that can be executed by a processor; wherein the processor is configured to execute the machine-executable instructions to implement any of the methods described in claims 12-16.

Images