Parameter optimization method and system for imaging structure of XRT ore sorting machine

Abstract

The invention discloses a parameter optimization method and system for an imaging structure of an XRT ore sorting machine. The parameter optimization method comprises the steps that structure parameter information of a target XRT ore sorting machine is obtained; constructing a simulation model of the target XRT ore sorting machine in simulation software; the influence of each element in the X-ray imaging structure on the signal-to-noise ratio of the image is researched, and a final influence factor is determined; and by taking the maximum signal-to-noise ratio and the minimum X-ray source power as optimization objectives and taking actual working conditions and process conditions as constraints, constructing a parameter optimization model of the imaging structure of the XRT ore sorting machine and solving the parameter optimization model to finish parameter optimization of the imaging structure of the target XRT ore sorting machine. According to the method, accurate modeling and simulation are conducted on the target XRT ore sorting machine, targeted parameter optimization model construction and solution are conducted after the influence factors of the parameters are determined, parameter optimization of the imaging structure of the target XRT ore sorting machine is achieved, and the scheme is relatively simple, higher in precision and better in effect.

Description

Technical Field

[0001] The present invention belongs to the technical field of mining equipment, and in particular relates to a parameter optimization method and system for an imaging structure of an XRT ore sorter. Background Art

[0002] The XRT ore sorter is an intelligent mineral processing equipment that uses XRT technology to sort ore. In the XRT ore sorter, the ore is evenly distributed at the beginning of the conveyor belt through vibrating feeders and baffle rubbers, and then transported to the position above the detection card via the conveyor belt. The ray source emits X-rays through the ore and irradiates the detection card to form an ore image. The ore image information is then transmitted to the signal processing center, which analyzes the ore grade and ore distribution position, controls the air nozzle to spray tailings, and completes the ore sorting. In the working process of the XRT ore sorter, the most critical ore imaging depends on the X-ray sensor detection system, which is the prerequisite for realizing ore sorting; and the imaging structure is one of the key components, which is responsible for the ore imaging part. The schematic diagram of the X-ray imaging structure is shown in the figure below. Figure 1 As shown in the figure, the numbers are: 1 is the lead cylinder, 2 is the lead-antimony collimator, 3 is the protective box, 4 is the protective partition, 5 is the lead plate under the belt, 6 is the detection card, 7 is the tungsten target, and 8 is the beryllium filter. The imaging principle is: the conveyor belt transports the ore to the area to be detected, and the X-rays generated by the electron bombardment of the tungsten target penetrate the ore and irradiate the detector. The detector will capture the X-ray signal and convert it into an electrical signal to form an ore image.

[0003] The signal-to-noise ratio (SNR) of an X-ray image is a key indicator of image quality. It reflects the detectability of valid information within the image amidst noise. A high SNR means more valid information can be identified within the image, enabling more accurate identification of useful minerals and waste rock, thereby effectively improving the sorting efficiency and accuracy of the XRT ore sorter.

[0004] Currently, existing research results primarily focus on X-ray image processing solutions and methods for increasing X-ray intensity. X-ray image processing solutions typically utilize various algorithms (such as numerical simulation, wavelet transform, transient noise model, and coherence analysis) to process the resulting images to achieve a higher signal-to-noise ratio. However, these solutions are often extremely complex. While seemingly simple, increasing X-ray intensity can effectively attenuate X-rays in the ore at excessive intensity, making the information carried difficult to discern. Furthermore, excessively high X-ray sources can lead to excessive X-ray leakage, increasing the radiation protection costs of XRT ore sorting machines. Summary of the Invention

[0005] One of the objectives of the present invention is to provide a parameter optimization method for an XRT ore sorting machine imaging structure that has a relatively simple solution, higher precision and better effect.

[0006] A second object of the present invention is to provide a system for implementing the parameter optimization method of the imaging structure of the XRT ore sorting machine.

[0007] The parameter optimization method of the imaging structure of the XRT ore sorter provided by the present invention comprises the following steps:

[0008] S1. Obtaining structural parameter information of the target XRT ore sorter;

[0009] S2. Based on the structural parameter information obtained in step S1, a simulation model of the target XRT ore sorter is constructed in the simulation software by combining the Monte Carlo simulation scheme, the Boltzmann equation, the cross-validation scheme, and the polynomial fitting scheme;

[0010] S3. Based on the simulation model obtained in step S2, a single factor analysis was performed based on the control variable method to study the effect of each element in the X-ray imaging structure on the image signal-to-noise ratio and determine the final influencing factors;

[0011] S4. Based on the influencing factors obtained in step S3, with maximum signal-to-noise ratio and minimum X-ray source power as the optimization goals, and with actual operating conditions and process conditions as constraints, a parameter optimization model for the imaging structure of the XRT ore sorter is constructed;

[0012] S5. Solve the model constructed in step S4 based on a genetic algorithm to optimize the parameters of the imaging structure of the target XRT ore sorter.

[0013] The step S2 comprises the following steps:

[0014] S2.1. Based on the structural parameter information obtained in step S1, a three-dimensional structural simulation model of the target XRT ore sorter is constructed in the simulation software;

[0015] S2.2. Simulate the propagation and interaction of X-rays from the target XRT ore sorter using a Monte Carlo simulation scheme and the Boltzmann equation.

[0016] S2.3. Based on a cross-validation scheme and a polynomial fitting scheme, a mathematical relationship between the energy and number of X-rays and the image signal-to-noise ratio is established.

[0017] The step S2.1 specifically includes the following steps:

[0018] Taking the X-ray imaging structure of the target XRT ore sorter as the research object, and taking the ray emission point as the origin, the Geant4 simulation software is used to establish the X-ray imaging structure simulation model of the target XRT ore sorter;

[0019] The constructed X-ray imaging structure simulation model includes a lead cylinder, a lead-antimony collimator, a protective box and protective partitions, a lead plate under the belt, and a scintillator of the detection card.

[0020] The step S2.2 specifically includes the following steps:

[0021] Based on the Monte Carlo simulation scheme, a set of state parameters is set for each particle to record all the current motion states of the particle; the state parameters include the spatial coordinates of the particle ,speed , the unit vector corresponding to the direction of motion , particle weight value and time ;

[0022] Ignoring the interaction between particles, the Boltzmann equation is used to record the particle motion process, which can be expressed as:

[0023] In the formula is the particle distribution function; is the spatial gradient of the particle distribution function; For particles in And the speed is Total cross section when is the velocity of the particle before collision; is the direction of motion of the particle before collision; is the collision time; is the scattering function, which is used to represent a velocity at position r. And the direction is Particles, after collision time After being scattered to the speed And the direction is The conditional probability within the unit speed and unit solid angle; For particles in And the speed is The total particle scattering cross section when ; is the source term, used to represent the time , at position At the unit time, unit volume, unit solid angle, the speed and direction The number of particles;

[0024] Based on the Monte Carlo simulation scheme, the process of simulating the particle output problem includes the following steps:

[0025] (1) Using normalized source term distribution Sampling is performed to determine the initial state of the particle ; And use the following formula to calculate the initial weight of the particle for ;in, is the total volume of the computational domain;

[0026] (2) Calculate the particle collision coordinates: Use the following formula to calculate the distance of particle movement :

[0027] In the formula Indicates that the particle is in the state Total cross section when To solve the particle movement distance The integral variable in the integral formula of ; is the free path of particle transport;

[0028] Calculate the particle collision coordinates for ;

[0029] (3) Determine whether the particle is in the simulation area: Set the particle's running area to ; When the particle leaves the region When the simulation stops; if the particle is in the region , then continue the simulation;

[0030] (4) Calculate particle collision time for ;

[0031] (5) Use the following formula to update the cumulative value of the target physical quantity:

[0032] In the formula is the current accumulated target physical quantity; is the target physical quantity accumulated previously; is the set global scale factor; is the weight function of the particle;

[0033] (6) Calculate the velocity and corresponding unit direction vector of the particle after collision, which are used to iteratively update the particles;

[0034] The scattering effect occurs after the particles collide. Distribution sampling confirmation;

[0035] Calculate the time it takes for particles to undergo scattering effects for , to achieve iterative update of particle state; among them, is the delay time after the collision, and according to Distribution sampling determination;

[0036] (7) The particle weight after the scattering effect is calculated using the following formula:

[0037] In the formula is the particle weight after the scattering effect occurs; is the scattering cross section; is the total cross section after scattering effect occurs;

[0038] Update the state parameters of the particles;

[0039] (8) Repeat steps (2) to (7) for iterative calculation until the number of calculated particles reaches the total number of particles; end the iterative calculation.

[0040] The step S2.3 specifically includes the following steps:

[0041] Under the same parameter conditions, several sets of data on average X-ray energy, number of X-rays and image signal-to-noise ratio were obtained through simulation;

[0042] Based on several sets of data obtained, a five-fold cross-validation scheme was used to obtain the difference between the training error and the validation error at the first-order, second-order, and third-order polynomials;

[0043] The polynomial fitting scheme is used for fitting, and the following formula is used as the final fitting function:

[0044] In the formula is the image signal-to-noise ratio; is the first fitting parameter; is the average energy of X-rays; is the second fitting parameter; is the number of X-rays; is the third fitting parameter; is the fourth fitting parameter; is the fifth fitting parameter.

[0045] The step S3 specifically includes the following steps:

[0046] According to the simulation model obtained in step S2, a single factor analysis is performed based on the control variable method to study the influence of each pixel in the X-ray imaging structure on the image signal-to-noise ratio, and the final influencing factors are determined, including the tube voltage. , tube current Targeting angle , beryllium filter thickness 、Lead tube opening width , lead antimony collimator opening width , opening width of protective partition and the width of the lead plate opening under the belt .

[0047] The step S4 specifically includes the following steps:

[0048] The following formula is used as the signal-to-noise ratio sub-objective function :

[0049] In the formula is a constant value; is the first weight value; is the second weight value; is the third weight value; is the value of the i-th influencing factor; is the value of the jth influencing factor; Corresponding to the tube voltage , tube current Targeting angle , beryllium filter thickness 、Lead tube opening width , lead antimony collimator opening width , opening width of protective partition and the width of the lead plate opening under the belt The value of

[0050] The following formula is used as the X-ray source power sub-objective function :

[0051] Taking the maximum signal-to-noise ratio and the minimum X-ray source power as the optimization goals, the objective function is constructed as follows: ;

[0052] Taking the actual working conditions and process conditions as constraints, the constraints are constructed and expressed as:

[0053] In the formula is the minimum value of the i-th influencing factor set according to actual working conditions and process conditions; is the maximum value of the i-th influencing factor set according to actual working conditions and process conditions.

[0054] The step S5 specifically includes the following steps:

[0055] The NSGA-Ⅱ algorithm is used to solve the model constructed in step S4 according to the obtained Pareto curve, and the parameter optimization of the imaging structure of the target XRT ore sorter is completed.

[0056] The present invention also provides a system for realizing the parameter optimization method of the imaging structure of the XRT ore sorting machine, comprising a structure parameter acquisition module, a simulation model construction module, an influencing factor determination module, an optimization module construction module and a parameter optimization module; the structure parameter acquisition module, the simulation model construction module, the influencing factor determination module, the optimization module construction module and the parameter optimization module are sequentially connected in series; the structure parameter acquisition module is used to acquire the structure parameter information of the target XRT ore sorting machine, and upload the data information to the simulation model construction module; the simulation model construction module is used to construct a simulation model of the target XRT ore sorting machine in the simulation software according to the received data information and the acquired structure parameter information, in combination with the Monte Carlo simulation scheme, the Boltzmann equation, the cross-validation scheme and the polynomial fitting scheme, and upload the data information to the simulation model construction module. Upload the influencing factor determination module; the influencing factor determination module is used to perform single factor analysis based on the control variable method according to the received data information and the obtained simulation model, study the influence of each element in the X-ray imaging structure on the image signal-to-noise ratio, and determine the final influencing factors, and upload the data information to the optimization module construction module; the optimization module construction module is used to construct a parameter optimization model of the XRT ore sorting machine imaging structure according to the received data information and the obtained influencing factors, with the maximum signal-to-noise ratio and the minimum X-ray source power as the optimization goals, and with the actual working conditions and process conditions as constraints, and upload the data information to the parameter optimization module; the parameter optimization module is used to solve the constructed model based on the received data information and the genetic algorithm to complete the parameter optimization of the target XRT ore sorting machine imaging structure.

[0057] The parameter optimization method and system for the imaging structure of an XRT ore sorter provided by the present invention not only achieves parameter optimization of the imaging structure of the target XRT ore sorter by accurately modeling and simulating the target XRT ore sorter and constructing and solving a targeted parameter optimization model after determining the influencing factors of the parameters, but also the solution of the present invention is relatively simple, has higher accuracy and better effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 The figure is a schematic diagram of the X-ray imaging structure of an existing XRT ore sorting machine; wherein, Figure 1 (a) is a three-dimensional schematic diagram of the front view of the X-ray imaging structure. Figure 1 (b) is a side 3D cross-sectional view of the X-ray imaging structure.

[0059] Figure 2Schematic diagram of the process of the present invention.

[0060] Figure 3 Schematic diagram of the proportions of the three effects of the embodiment of the method of the present invention under different X-ray energies and different atomic numbers of substances.

[0061] Figure 4 Schematic diagram of the curves showing the changes in training error and verification error at different orders of the method embodiment of the present invention.

[0062] Figure 5 Schematic diagram of the fitting result surface of the embodiment of the method of the present invention.

[0063] Figure 6 Schematic diagram of residual distribution of an embodiment of the method of the present invention.

[0064] Figure 7 Schematic diagram of the signal-to-noise ratio variation under different parameters of the method embodiment of the present invention; wherein, Figure 7 (a) is a schematic diagram of the signal-to-noise ratio change curve under different tube voltages. Figure 7 (b) is a schematic diagram of the signal-to-noise ratio change curve under different tube currents. Figure 7 (c) is a schematic diagram of the signal-to-noise ratio change curve under different targeting angles. Figure 7 (d) is a schematic diagram of the signal-to-noise ratio change curve under different beryllium filter thicknesses. Figure 7 (e) is a schematic diagram of the signal-to-noise ratio change curve under different lead cylinder opening widths. Figure 7 (f) is a schematic diagram of the signal-to-noise ratio change curve under different lead-antimony collimator opening widths. Figure 7 (g) is a schematic diagram of the signal-to-noise ratio change curve under different protective partition opening widths. Figure 7 (h) is a schematic diagram of the signal-to-noise ratio change curve under different lead plate opening widths under the belt.

[0065] Figure 8 Schematic diagram of the curve showing the Pareto frontier changing with the number of iterations in an embodiment of the method of the present invention; wherein, Figure 8 (a) is a schematic diagram of the curve of 1 iteration, Figure 8 (b) is a schematic diagram of the curve of 10 iterations. Figure 8 (c) is a schematic diagram of the curve of 50 iterations. Figure 8 (d) is a schematic diagram of the curve after 100 iterations.

[0066] Figure 9 Schematic diagram of image signal-to-noise ratio change curves before and after optimization under different ray source powers in an embodiment of the method of the present invention.

[0067] Figure 10 Schematic diagram of the functional modules of the system of the present invention. DETAILED DESCRIPTION

[0068] like Figure 2 The figure shows a schematic flow chart of the method of the present invention: the parameter optimization method of the imaging structure of the XRT ore sorter disclosed in the present invention comprises the following steps:

[0069] S1. Obtain the structural parameter information of the target XRT ore sorter.

[0070] S2. Based on the structural parameter information obtained in step S1, a simulation model of the target XRT ore sorter is constructed in the simulation software by combining the Monte Carlo simulation scheme, the Boltzmann equation, the cross-validation scheme, and the polynomial fitting scheme.

[0071] The specific implementation includes the following steps:

[0072] S2.1. Based on the structural parameter information obtained in step S1, construct a three-dimensional structural simulation model of the target XRT ore sorter in the simulation software; specifically, the following steps are included:

[0073] Taking the X-ray imaging structure of the target XRT ore sorter as the research object, and taking the ray emission point as the origin, the Geant4 simulation software is used to establish the X-ray imaging structure simulation model of the target XRT ore sorter;

[0074] The constructed X-ray imaging structure simulation model includes a lead cylinder, a lead-antimony collimator, a protective box and protective partitions, a lead plate under the belt, and a scintillator of the detection card.

[0075] S2.2. Simulate the propagation and interaction of X-rays from the target XRT ore sorter using a Monte Carlo simulation scheme and the Boltzmann equation. The simulation steps are as follows:

[0076] In the simulation of X-ray imaging structures, electrons originate from the cathode of the X-ray source, are accelerated by the electric field of the tube voltage, and bombard the tungsten target, generating continuous X-rays and characteristic X-rays. In the subsequent imaging process, the X-rays interact with the imaging structure. These are general issues in the particle transport process, so the Monte Carlo method is used to simulate the propagation process and interaction of electrons and X-rays in the X-ray imaging structure.

[0077] Based on the Monte Carlo simulation scheme, a set of state parameters is set for each particle to record all the current motion states of the particle; the state parameters include the spatial coordinates of the particle ,speed , the unit vector corresponding to the direction of motion , particle weight value and time ;

[0078] Ignoring the interaction between particles, the Boltzmann equation is used to record the particle motion process, which can be expressed as:

[0079] In the formula is the particle distribution function; is the spatial gradient of the particle distribution function; For particles in And the speed is Total cross section when is the velocity of the particle before collision; is the direction of motion of the particle before collision; is the collision time; is the scattering function, which is used to represent a velocity at position r. And the direction is Particles, after collision time After being scattered to the speed And the direction is The conditional probability within the unit speed and unit solid angle; For particles in And the speed is The total particle scattering cross section when ; is the source term, used to represent the time , at position At the unit time, unit volume, unit solid angle, the speed and direction The number of particles;

[0080] Based on the Monte Carlo simulation scheme, the process of simulating the particle output problem includes the following steps:

[0081] (1) Using normalized source term distribution Sampling is performed to determine the initial state of the particle ; And use the following formula to calculate the initial weight of the particle for ;in, is the total volume of the computational domain;

[0082] (2) Calculate the particle collision coordinates: Use the following formula to calculate the distance of particle movement :

[0083] In the formula Indicates that the particle is in the state Total cross section when is the integral variable; is the free path of particle transport;

[0084] Calculate the particle collision coordinates for ;

[0085] (3) Determine whether the particle is in the simulation area: Set the particle's running area to ; When the particle leaves the region When the simulation stops; if the particle is in the region , then continue the simulation;

[0086] (4) Calculate particle collision time for ;

[0087] (5) Use the following formula to update the cumulative value of the target physical quantity:

[0088] In the formula is the current accumulated target physical quantity; is the target physical quantity accumulated previously; is the set global scale factor (related to the system geometry and simulation framework); is the weight function of the particle;

[0089] (6) Calculate the velocity and corresponding unit direction vector of the particle after collision, which are used to iteratively update the particles;

[0090] After the particles collide, there is a scattering effect, and the speed There has been a change, Distribution sampling confirmation;

[0091] Calculate the time it takes for particles to undergo scattering effects for , to achieve iterative update of particle state; among them, is the delay time after the collision, and according to Distribution sampling determination;

[0092] (7) The particle weight after the scattering effect is calculated using the following formula:

[0093] In the formula is the particle weight after the scattering effect occurs; is the scattering cross section; is the total cross section after scattering effect occurs;

[0094] Update the state parameters of the particles;

[0095] (8) Repeat steps (2) to (7) for iterative calculation until the number of calculated particles reaches the total number of particles; end the iterative calculation;

[0096] At the end of the simulation model, you need to set up a method for collecting particle data during the simulation. You need to define the entire process from the initial emission of particles to each particle collision event and finally to the particle decay. Data collection and processing must first determine the particle information, and then collect and process the particle information that needs to be counted. When a particle is first emitted, edit the particle initialization behavior in G4UserRunAction. Based on the core of Monte Carlo theory, edit the particle's random number generator to simulate the uncertainty in the physical process and ensure the accuracy of the simulation results. Particle information judgment and collection: Each collision event simulation requires editing and judgment in (G4UserEventAction). When the particle entering the detector card geometry is an X-ray, the X-ray energy and number are recorded based on the X-ray data recorded by the Monte Carlo method. The particle end information is edited and output in G4UserEventAction. The data processing adopts the root method, recording the energy of each X-ray entering the detector card, accumulating the X-ray number, and then outputting it as a histogram to obtain the X-ray energy distribution diagram, completing the establishment of the entire simulation model.

[0097] S2.3. Establish a mathematical relationship between X-ray energy and number and image signal-to-noise ratio based on a cross-validation and polynomial fitting scheme. This includes the following steps:

[0098] The model constructed by the Geant4 simulation software cannot simulate the subsequent circuit structure and cannot directly obtain the signal-to-noise ratio of the image. It can only obtain the energy and quantity of X-rays received by the detector card. In actual imaging, the detector card also receives the energy and intensity of X-rays and emits fluorescence. After the circuit structure converts the fluorescence into photoelectricity and transmits digital information, a digital image is finally formed. Therefore, to obtain the signal-to-noise ratio of the image, it is necessary to combine experiments with simulations. The X-ray energy obtained by the successfully verified X-ray imaging structure model and the image signal-to-noise ratio value obtained by experiment under the same parameters can be used to establish a mathematical relationship between X-ray energy and quantity and image signal-to-noise ratio, and then evaluate the difference in image quality caused by the optimized imaging structure parameters.

[0099] Under the same parameter conditions, several sets of data on average X-ray energy, number of X-rays and image signal-to-noise ratio were obtained through simulation;

[0100] Based on several sets of data obtained, a five-fold cross-validation scheme was used to obtain the difference between the training error and the validation error at the first-order, second-order, and third-order polynomials;

[0101] The polynomial fitting scheme is used for fitting, and the following formula is used as the final fitting function:

[0102] In the formula is the image signal-to-noise ratio; is the first fitting parameter; is the average energy of X-rays; is the second fitting parameter; is the number of X-rays; is the third fitting parameter; is the fourth fitting parameter; is the fifth fitting parameter.

[0103] S3. Based on the simulation model obtained in step S2, a single factor analysis is performed based on the control variable method to study the impact of each element in the X-ray imaging structure on the image signal-to-noise ratio and determine the final influencing factors; specifically, the following steps are included:

[0104] According to the simulation model obtained in step S2, a single factor analysis is performed based on the control variable method to study the influence of each pixel in the X-ray imaging structure on the image signal-to-noise ratio and determine the final influencing factors; the final influencing factors determined include tube voltage , tube current Targeting angle , beryllium filter thickness 、Lead tube opening width , lead antimony collimator opening width , opening width of protective partition and the width of the lead plate opening under the belt .

[0105] S4. Based on the influencing factors obtained in step S3, with maximizing the signal-to-noise ratio and minimizing the X-ray source power as the optimization goals, and subject to actual operating conditions and process conditions, a parameter optimization model for the imaging structure of the XRT ore sorter is constructed. Specifically, the model comprises the following steps:

[0106] The following formula is used as the signal-to-noise ratio sub-objective function :

[0107] In the formula is a constant value; is the first weight value; is the second weight value; is the third weight value; is the value of the i-th influencing factor; is the value of the jth influencing factor; Corresponding to the tube voltage , tube current Targeting angle , beryllium filter thickness 、Lead tube opening width , lead antimony collimator opening width , opening width of protective partition and the width of the lead plate opening under the belt The value of

[0108] The following formula is used as the X-ray source power sub-objective function :

[0109] Taking the maximum signal-to-noise ratio and the minimum X-ray source power as the optimization goals, the objective function is constructed as follows: ;

[0110] Taking the actual working conditions and process conditions as constraints, the constraints are constructed and expressed as:

[0111] In the formula is the minimum value of the i-th influencing factor set according to actual working conditions and process conditions; is the maximum value of the i-th influencing factor set according to actual working conditions and process conditions.

[0112] S5. Based on the genetic algorithm, the model constructed in step S4 is solved to complete the parameter optimization of the target XRT ore sorting machine imaging structure; specifically comprising the following steps:

[0113] The solution to multi-objective optimization is not a single solution, but a set of optimal solutions, where several objective functions reach a balance. These solutions form a Pareto curve. Solutions outside this set will compromise the values ​​of other objective functions while optimizing another objective function. Therefore, the specific selection of the optimal solution for multi-objective optimization needs to be based on the Pareto curve according to the actual project.

[0114] There are many strategies for solving multi-objective optimization models. Common strategies include weighted methods, ideal point methods, NSGA-II, and multi-objective particle swarm optimization. Both NSGA-II and multi-objective particle swarm optimization are effective methods for solving multi-objective optimization problems and obtaining Pareto frontier solutions. However, compared with multi-objective particle swarm optimization, NSGA-II can explore a wider search space, has stronger adaptability, can maintain solution diversity, avoid convergence to local optimal solutions, and is suitable for finding a uniform distribution on the Pareto frontier. Therefore, NSGA-II was chosen.

[0115] Therefore, the NSGA-Ⅱ algorithm is finally used to solve the model constructed in step S4 according to the obtained Pareto curve to complete the parameter optimization of the imaging structure of the target XRT ore sorter.

[0116] The method of the present invention is further described below with reference to an embodiment:

[0117] Taking the X-ray imaging structure of 1200 XRT ore sorting machine as the research object, a simulation model was first established in Geant4 simulation software.

[0118] The X-ray source used in the 1200 XRT ore sorter is a Gaomei integrated source. The electron emission energy and quantity are set according to the source's tube voltage and current, with the electron emission direction maintaining a 25° angle with the target surface. The X-ray tube inside the Gaomei source is made of cermet, which has excellent high-temperature resistance and thermal conductivity, capable of withstanding the high heat generated by X-rays and meeting their heat dissipation requirements. A vacuum environment is maintained inside the tube to prevent electron attenuation during airborne propagation, which would waste energy and prevent filament burnout. X-ray tube parameters are shown in Table 1:

[0119] The lead tube is an important component of the X-ray source. Its function is to shield and encapsulate the X-ray tube, prevent X-rays from leaking in unnecessary directions, and ensure the mechanical stability of the X-ray source and the directionality of the rays. The parameter values ​​of the lead tube geometric model are shown in Table 2.

[0120] The opening width of the lead-antimony collimator is used to collimate the X-rays, forming a parallel beam with strong directionality and small divergence, avoiding interference from stray rays, controlling the direction of the X-ray beam, and ensuring a clear image in the detection area. The shape of the lead-antimony collimator is a rectangular parallelepiped, and its modeling parameter values ​​are shown in Table 3. The shape of the lead-antimony opening is a trapezoid, and its modeling parameter values ​​are shown in Table 4.

[0121] The protective box and protective partition are used to locally shield and isolate some areas in the system to prevent X-rays from leaking to non-target areas. The protective box has X-rays penetrating from the lead antimony collimator. The opening width of the protective partition can control the range and direction of the ray beam when the rays pass through. The modeling parameter values ​​of the protective box and protective partition are shown in Table 5.

[0122] The lead plate under the belt provides mechanical support and radiation protection for the detection belt. The detection belt is used to transport ore. The opening size is used to reduce the scattering of X-rays while allowing the X-rays that penetrate the ore to enter the detection card. The modeling parameter values ​​of the lead plate under the belt are shown in Table 6.

[0123] The detection card is used to collect X-rays that penetrate the ore and convert them into electrical signals. After subsequent processing, a grayscale image of the ore is formed. The part of the detection card that receives X-rays and emits fluorescence is the scintillator. Therefore, when modeling, only the scintillator model needs to be constructed. The modeling parameter values ​​of the scintillator are shown in Table 7.

[0124] The above are geometric modeling diagrams of various parts of the X-ray imaging structure, which affect the entire X-ray propagation process and interaction with matter, affect the amount of information carried by the image, cause changes in the image signal-to-noise ratio, and change the sorting accuracy and efficiency of the XRT ore sorter.

[0125] There are three main types of interactions between X-rays and matter: photoelectric effect, Compton effect, and electron pair effect. They behave differently under different energies and atomic numbers of matter, such as Figure 3 As shown: When the energy is MeV, the photoelectric effect dominates; at energies MeV, the Compton effect dominates; at energies At MeV, the electron pair effect dominates. In the imaging structure of an XRT ore sorter, the tube voltage of the X-ray source ranges from 100 to 180 kV, and the generated X-ray energy ranges from 100 to 180 keV. Therefore, the main X-ray photoelectric effect and Compton effect occur in the physical process of the imaging structure.

[0126] Then, the particle motion process is simulated based on the Monte Carlo method.

[0127] Determine the factors affecting the image signal-to-noise ratio:

[0128] Experiments were conducted to obtain the average X-ray energy and number of X-rays obtained by simulation under the same parameter conditions, as well as the image signal-to-noise ratio obtained by experiment, as shown in Tables 8 and 9:

[0129] In order to determine the best fitting mathematical function between the average X-ray energy E, the number of X-rays N and the signal-to-noise ratio SNR; the five-fold cross-validation technique was used to obtain the difference between the training error and the validation error when the first-order, second-order and third-order polynomials were used; the results were implemented through MATLAB code. Figure 4 As shown in the figure, when the second-order polynomial is used, the mean square error of the training set is closest to the mean square error of the validation set, so the second-order polynomial is selected as the fitting function.

[0130] Because when the average energy and number of X-rays are 0, the signal-to-noise ratio is also 0, so the signal-to-noise ratio fitting function must pass through the origin. Then the fitting function has no constant term and can be expressed as:

[0131] Finally, the fitting results are as follows Figure 5 As shown, the fitting function is shown in the following formula, and the goodness of fit R2 of this equation is 0.991.

[0132] In order to verify whether the fitting function is reasonable, the residual distribution diagram is drawn according to the fitting function, as shown in Figure 6 As shown in the figure, the residuals are evenly distributed on both sides of the zero line, without showing any specific structure or pattern, and all the residual points are distributed within the range of 3δ, without any abnormal points, indicating that the residuals are randomly distributed and the fitting function is reasonable; therefore, the obtained fitting formula can be used to calculate the image signal-to-noise ratio by the average energy and number of X-rays, and more intuitively judge the influence of various parameters of the imaging structure on the image signal-to-noise ratio.

[0133] Univariate analysis was performed to determine the final influencing factors:

[0134] The influence of X-ray imaging structure on image signal-to-noise ratio is reflected in two aspects. One is that in the stage of generating X-rays, the parameters of the X-ray source will directly determine the energy distribution of the X-rays after generation. The other is that before the X-rays propagate to the detection card, each imaging structure limits the energy distribution range of the X-rays and produces attenuation. The factors of the X-ray source are tube voltage, tube current, and target angle. These three factors directly affect the energy intensity of the generated X-rays and the emission direction of the X-rays, thereby affecting the subsequent image signal-to-noise ratio. The beryllium filter will screen out X-rays with too low energy and enhance the overall average energy of the X-rays, but it will also affect the number of X-rays and the final image signal. Noise ratio; the opening width of the lead cylinder, lead antimony collimator, protective baffle, and lead plate under the belt will limit the energy distribution of X-rays and reduce the leakage radiation and scattered radiation of X-rays. On the other hand, it will reduce the utilization rate of X-rays due to excessive protection, waste some X-rays, and reduce the signal-to-noise ratio of the image; therefore, eight factors such as tube voltage, tube current, target angle, lead cylinder opening width, beryllium filter thickness, lead antimony collimator opening width, protective baffle opening width, and lead plate opening width under the belt are selected to study the influence of X-ray imaging structure on image signal-to-noise ratio, and the control variable method is used to study the influence of various factors of X-ray imaging structure on image signal-to-noise ratio. Draw an image such as Figure 7 shown.

[0135] Figure 7 (a) shows that the signal-to-noise ratio increases significantly with increasing tube voltage. Increasing tube voltage increases the upper energy limit of the emitted electrons, raising the upper energy limit of X-rays produced by bremsstrahlung radiation and increasing the number of characteristic X-rays produced by characteristic radiation, thereby increasing the overall energy of X-rays. Simultaneously, increasing tube voltage increases the kinetic energy of electrons emitted by the cathode in the X-ray source, leading to more bremsstrahlung and characteristic radiation from collisions between electrons and the anode tungsten target, thus increasing the number of X-rays produced.

[0136] Figure 7 (b) shows that the signal-to-noise ratio significantly improves with increasing tube current. Increasing tube current increases the number of electrons emitted by the X-ray tube cathode, generating more bremsstrahlung and characteristic radiation, and increasing the number of continuous and characteristic X-rays. This increases the total number of X-rays, the number of X-rays received by the detector card, and the signal strength.

[0137] Figure 7Figure (c) shows that within the 5-60° range, the signal-to-noise ratio (SNR) increases with increasing targeting angle, reaching a peak at 60°. Then, as the targeting angle increases, the SNR gradually decreases. As the targeting angle increases from 0° to 60°, the electron beam's active area on the target decreases, concentrating the effective focal area. This enhances the directionality of the X-rays, increases the number of X-rays reaching the detector card, and increases the proportion of high-energy X-rays, improving signal strength and increasing the SNR. As the targeting angle increases from 60° to 90°, the electron beam's effective bombardment area on the target decreases rapidly, increasing the collision depth between the electrons and the target. Some of the energy is wasted in unproductive heat loss. Furthermore, the X-rays generated at large angles have poor directionality, generating more low-energy X-rays and scattered radiation, increasing the impact of background noise. Under these dual effects, the SNR gradually decreases. As the targeting angle approaches 90°, the target's geometry hinders the release of some X-rays, further weakening the effective signal.

[0138] Figure 7 (d) shows that the signal-to-noise ratio decreases as the beryllium filter thickness decreases. When the beryllium filter thickness is greater than 2.6 mm, the signal-to-noise ratio no longer decreases with the increase in beryllium filter thickness. This is because when the beryllium filter thickness increases to 2.6 mm, excessive low-energy X-rays have been effectively filtered out, and the source of the signal-to-noise ratio signal is basically high-energy X-rays. The beryllium filter cannot effectively filter out high-energy X-rays, so the signal intensity no longer changes and the signal-to-noise ratio remains unchanged.

[0139] Figure 7 (e) shows that: in the range of 0-35mm, the signal-to-noise ratio increases with the increase in the lead cylinder opening width; the signal-to-noise ratio reaches its peak when the lead cylinder opening width increases to 35mm; and the signal-to-noise ratio remains unchanged as the opening width increases beyond 35mm. When the lead cylinder opening width is between 0-35mm, the increase in the lead cylinder opening width increases the number of X-rays passing through, increasing the number of X-rays received by the detector card, enhancing the signal strength, and increasing the image signal-to-noise ratio; when the lead cylinder opening width is between 35mm and 45mm, the number of X-rays received by the detector card reaches saturation, the signal-to-noise ratio reaches its peak, and an optimal balance is achieved between signal and noise; subsequent increases in the opening width may increase the total number of X-rays passing through the lead cylinder, but the number of X-rays within the X-ray distribution range received by the detector card remains unchanged, the number of received X-rays reaches saturation, the signal strength stabilizes, and the signal-to-noise ratio remains unchanged.

[0140] Figure 7(f) shows that when the lead antimony collimator opening width is 5-25mm, the signal-to-noise ratio increases with the increase of the lead antimony collimator opening width. When the lead antimony collimator opening width is 25mm-30mm, the signal-to-noise ratio no longer increases with the increase of the lead antimony collimator opening width. Increasing the lead antimony collimator opening width reduces the restriction and absorption of X-rays, resulting in an increase in the number of X-rays within the X-ray distribution range that the detector card can receive, an increase in signal strength, and an increase in the signal-to-noise ratio, reaching a peak at 25mm. When the opening width is between 25mm and 30mm, the total number of X-rays passing through the collimator continues to increase, but within the detection card's acceptance range, the number of X-rays no longer changes, and the signal strength remains unchanged. Therefore, the signal-to-noise ratio tends to stabilize after the lead antimony collimator opening width exceeds 25mm.

[0141] Figure 7 (g) shows that when the opening width of the protective baffle is between 1 and 4 mm, the signal-to-noise ratio increases significantly with the increase of the opening width. When the opening width of the protective baffle is between 4 and 9 mm, the signal-to-noise ratio increases slowly with the increase of the opening width. When the opening width is greater than 9 mm, the signal-to-noise ratio no longer increases. When the opening width gradually increases from 1 mm to 4 mm, the number of X-rays passing through the protective baffle increases significantly, and the number of X-rays within the detection card's receiving range increases significantly, which directly increases the number of X-rays reaching the detector, significantly enhancing the signal strength and the signal-to-noise ratio. When the opening width continues to increase to 4-9 mm, the total number of X-rays passing through the protective baffle increases, but the number of X-rays within the detection card's receiving range only increases slightly, approaching saturation, the signal strength slowly increases, and the signal-to-noise ratio increases slightly to remain unchanged. When the opening width is greater than 9 mm, the total number of X-rays continues to increase, but within the detection card's receiving range, the number of X-rays no longer changes, the signal strength does not increase, and the signal-to-noise ratio remains unchanged.

[0142] Figure 7 Figure (h) shows that when the opening width of the lead plate under the belt is between 1 and 2.4 mm, the signal-to-noise ratio increases with the increase in the opening width. When the opening width exceeds 2.4 mm, the signal-to-noise ratio remains constant. As the opening width increases from 1 mm to 2.4 mm, the number of X-rays passing through the lead plate under the belt increases, and the number of X-rays within the detection card's reception range increases, improving the detector's signal strength and the signal-to-noise ratio. When the opening width exceeds 2.4 mm, the number of X-rays passing through the lead plate under the belt and irradiating the detection card no longer increases, nor does the signal strength change, and the signal-to-noise ratio remains stable.

[0143] Model construction and solution:

[0144] There are two objectives in optimizing the X-ray imaging structure of an XRT ore sorter. On the one hand, the SNR of the X-ray imaging structure is improved to improve the sorter's recognition accuracy, with the maximum SNR as the optimization goal. On the other hand, the efficiency of the X-ray source is improved by reducing the X-ray source power, with the minimum X-ray source power as the optimization goal. When constructing the optimization objectives, the tube voltage (U), tube current (I), target angle (α), beryllium filter thickness (d), lead cylinder opening width (k1), lead antimony collimator opening width (k2), protective baffle opening width (k3), and lead plate opening width under the belt (k4) are used as decision variables. The optimization objectives are to maximize the SNR and minimize the X-ray source power. The constraints are limited by actual operating conditions and process conditions, as shown in Table 10:

[0145] The image signal-to-noise ratio of XRT ore sorting determines the recognition accuracy of the XRT ore sorting, thus affecting the sorting performance of the entire machine. A too low image signal-to-noise ratio will make it difficult for the machine to accurately identify the grade of the ore, affecting the sorting ability of the sorting machine. The objective function is constructed with the maximum signal-to-noise ratio as the goal. However, the NSGA-II algorithm takes the minimum objective function value as the optimization goal, so the original maximum objective function is converted to the minimum value by adding a negative sign to it. The objective function is:

[0146] The X-ray source power of the XRT ore sorter determines the X-ray emission intensity. Excessive X-ray source power increases the risk of X-ray radiation leakage and the cost of X-ray protection, thereby reducing the economic benefits of the XRT ore sorter. The objective function is constructed with the goal of minimizing the X-ray source power: ;

[0147] Combining the above objective functions, structural parameters, and constraints, the multi-objective optimization model of the XRT ore sorter X-ray imaging structure is as follows:

[0148] The multi-objective optimization problem of X-ray imaging structures was solved by NSGA-II. The algorithm was implemented with the help of MATLAB and the Pareto curve was plotted. The core parameters of the algorithm include population size, crossover probability, mutation probability, and number of iterations. The parameter settings are shown in Table 11:

[0149] Solve the multi-objective optimization model of X-ray imaging structure and draw the Pareto optimal solution set. The Pareto optimal solution set curve is different under different iteration numbers. When the iteration number is 100, the Pareto optimal solution set curve is stable, and the points in the optimal solution set are evenly distributed on the curve, such as Figure 8 shown.

[0150] In the optimization process of the NSGA-Ⅱ algorithm, the optimal solutions on the Pareto front were obtained after 100 iterations. These solutions are the result of the coordination of the two optimization objectives after the optimization of the X-ray imaging structure parameters. There is no distinction between good and bad. Researchers can choose the most suitable solution from them according to the actual working conditions.

[0151] According to experience, the X-ray source power of the XRT ore sorter should not exceed 800 W. Based on this, the maximum signal-to-noise ratio of the image before optimization was 75.89, and the maximum signal-to-noise ratio of the image after optimization was 94.26, an improvement of 24.21%. The specific parameters after optimization are shown in Table 12:

[0152] Under the same ray source power, the optimized results are significantly improved compared with those before optimization. Figure 9 This indicates that the X-ray source power is within the allowable range. By optimizing X-ray imaging structural parameters, the efficiency of X-ray source power utilization can be improved, significantly enhancing the image signal-to-noise ratio. XRT ore sorting machine engineers can select the appropriate X-ray source model based on the actual operating conditions and process requirements, reducing unnecessary power waste.

[0153] To ensure the credibility of the optimization scheme, a simulation model was set up based on the parameters in Table 12. After simulation calculation, the optimized signal-to-noise ratio value was obtained. It was compared with the actual optimization results, and the results are shown in Table 13:

[0154] It can be seen from Table 13 that the relative error between the simulation value and the optimized result is small, which proves that the optimization solution obtained by the NSGA-Ⅱ algorithm for solving the multi-objective optimization model of X-ray imaging structure has a high credibility.

[0155] like Figure 10The figure shows a schematic diagram of the functional modules of the system of the present invention: the system disclosed in the present invention for realizing the parameter optimization method of the imaging structure of the XRT ore sorting machine comprises a structure parameter acquisition module, a simulation model construction module, an influencing factor determination module, an optimization module construction module and a parameter optimization module; the structure parameter acquisition module, the simulation model construction module, the influencing factor determination module, the optimization module construction module and the parameter optimization module are connected in series in sequence; the structure parameter acquisition module is used to obtain the structure parameter information of the target XRT ore sorting machine and upload the data information to the simulation model construction module; the simulation model construction module is used to construct a simulation model of the target XRT ore sorting machine in the simulation software according to the received data information and the obtained structure parameter information, in combination with the Monte Carlo simulation scheme, the Boltzmann equation, the cross-validation scheme and the polynomial fitting scheme. model, and upload the data information to the influencing factor determination module; the influencing factor determination module is used to perform single factor analysis based on the control variable method according to the received data information and the obtained simulation model, study the influence of each element in the X-ray imaging structure on the image signal-to-noise ratio, and determine the final influencing factors, and upload the data information to the optimization module construction module; the optimization module construction module is used to construct a parameter optimization model of the XRT ore sorting machine imaging structure according to the received data information and the obtained influencing factors, with the maximum signal-to-noise ratio and the minimum X-ray source power as the optimization goals, and with the actual working conditions and process conditions as constraints, and upload the data information to the parameter optimization module; the parameter optimization module is used to solve the constructed model based on the genetic algorithm according to the received data information, and complete the parameter optimization of the target XRT ore sorting machine imaging structure.

Claims

1. A parameter optimization method for the imaging structure of an XRT ore sorter, characterized in that The steps include: S1. Obtaining structural parameter information of the target XRT ore sorter; S2. Based on the structural parameter information obtained in step S1, a simulation model of the target XRT ore sorter is constructed in the simulation software by combining the Monte Carlo simulation scheme, the Boltzmann equation, the cross-validation scheme, and the polynomial fitting scheme; S3. Based on the simulation model obtained in step S2, a single factor analysis was performed based on the control variable method to study the effect of each element in the X-ray imaging structure on the image signal-to-noise ratio and determine the final influencing factors; S4. Based on the influencing factors obtained in step S3, with maximum signal-to-noise ratio and minimum X-ray source power as the optimization goals, and with actual operating conditions and process conditions as constraints, a parameter optimization model for the imaging structure of the XRT ore sorter is constructed; S5. Solve the model constructed in step S4 based on a genetic algorithm to optimize the parameters of the imaging structure of the target XRT ore sorter.

2. The parameter optimization method of the imaging structure of the XRT ore sorter according to claim 1, characterized in that The step S2 comprises the following steps: S2.

1. Based on the structural parameter information obtained in step S1, a three-dimensional structural simulation model of the target XRT ore sorter is constructed in the simulation software; S2.

2. Simulate the propagation and interaction of X-rays from the target XRT ore sorter using a Monte Carlo simulation scheme and the Boltzmann equation. S2.

3. Based on a cross-validation scheme and a polynomial fitting scheme, a mathematical relationship between the energy and number of X-rays and the image signal-to-noise ratio is established.

3. The parameter optimization method of the imaging structure of the XRT ore separator according to claim 2, characterized in that The step S2.1 specifically includes the following steps: Taking the X-ray imaging structure of the target XRT ore sorter as the research object, and taking the ray emission point as the origin, the Geant4 simulation software is used to establish the X-ray imaging structure simulation model of the target XRT ore sorter; The constructed X-ray imaging structure simulation model includes a lead cylinder, a lead-antimony collimator, a protective box and protective partitions, a lead plate under the belt, and a scintillator of the detection card.

4. The parameter optimization method of the imaging structure of the XRT ore separator according to claim 3, characterized in that The step S2.2 specifically includes the following steps: Based on the Monte Carlo simulation scheme, a set of state parameters is set for each particle to record all the current motion states of the particle; the state parameters include the spatial coordinates of the particle ,speed , the unit vector corresponding to the direction of motion , particle weight value and time ; Ignoring the interaction between particles, the Boltzmann equation is used to record the particle motion process, which can be expressed as: In the formula is the particle distribution function; is the spatial gradient of the particle distribution function; For particles in And the speed is Total cross section when is the velocity of the particle before collision; is the direction of motion of the particle before collision; is the collision time; is the scattering function, which is used to represent a velocity at position r. And the direction is Particles, after collision time After being scattered to the speed And the direction is The conditional probability within the unit speed and unit solid angle; For particles in And the speed is The total particle scattering cross section when ; is the source term, used to represent the time , at position At the unit time, unit volume, unit solid angle, the speed and direction The number of particles; Based on the Monte Carlo simulation scheme, the process of simulating the particle output problem includes the following steps: (1) Using normalized source term distribution Sampling is performed to determine the initial state of the particle ; And use the following formula to calculate the initial weight of the particle for ;in, is the total volume of the computational domain; (2) Calculate the particle collision coordinates: Use the following formula to calculate the distance of particle movement : In the formula Indicates that the particle is in the state Total cross section when is the integral variable; is the free path of particle transport; Calculate the particle collision coordinates for ; (3) Determine whether the particle is in the simulation area: Set the particle's running area to ; When the particle leaves the region When the simulation stops; if the particle is in the region , then continue the simulation; (4) Calculate particle collision time for ; (5) Use the following formula to update the cumulative value of the target physical quantity: In the formula is the current accumulated target physical quantity; is the target physical quantity accumulated previously; is the set global scale factor; is the weight function of the particle; (6) Calculate the velocity and corresponding unit direction vector of the particle after collision, which are used to iteratively update the particles; The scattering effect occurs after the particles collide. Distribution sampling confirmation; Calculate the time it takes for particles to undergo scattering effects for , to achieve iterative update of particle state; among them, is the delay time after the collision, and according to Distribution sampling determination; (7) The particle weight after the scattering effect is calculated using the following formula: In the formula is the particle weight after the scattering effect occurs; is the scattering cross section; is the total cross section after scattering effect occurs; Update the state parameters of the particles; (8) Repeat steps (2) to (7) for iterative calculation until the number of calculated particles reaches the total number of particles; end the iterative calculation.

5. The parameter optimization method of the imaging structure of the XRT ore separator according to claim 4, characterized in that The step S2.3 specifically includes the following steps: Under the same parameter conditions, several sets of data on average X-ray energy, number of X-rays and image signal-to-noise ratio were obtained through simulation; Based on several sets of data obtained, a five-fold cross-validation scheme was used to obtain the difference between the training error and the validation error at the first-order, second-order, and third-order polynomials; The polynomial fitting scheme is used for fitting, and the following formula is used as the final fitting function: In the formula is the image signal-to-noise ratio; is the first fitting parameter; is the average energy of X-rays; is the second fitting parameter; is the number of X-rays; is the third fitting parameter; is the fourth fitting parameter; is the fifth fitting parameter.

6. The parameter optimization method of the imaging structure of the XRT ore separator according to claim 5, characterized in that The step S3 specifically includes the following steps: According to the simulation model obtained in step S2, a single factor analysis is performed based on the control variable method to study the influence of each pixel in the X-ray imaging structure on the image signal-to-noise ratio, and the final influencing factors are determined, including the tube voltage. , tube current Targeting angle , beryllium filter thickness 、Lead tube opening width , lead antimony collimator opening width , opening width of protective partition and the width of the lead plate opening under the belt .

7. The parameter optimization method of the imaging structure of the XRT ore separator according to claim 6, characterized in that The step S4 specifically includes the following steps: The following formula is used as the signal-to-noise ratio sub-objective function : In the formula is a constant value; is the first weight value; is the second weight value; is the third weight value; is the value of the i-th influencing factor; is the value of the jth influencing factor; Corresponding to the tube voltage , tube current Targeting angle , beryllium filter thickness 、Lead tube opening width , lead antimony collimator opening width , opening width of protective partition and the width of the lead plate opening under the belt The value of The following formula is used as the X-ray source power sub-objective function : Taking the maximum signal-to-noise ratio and the minimum X-ray source power as the optimization goals, the objective function is constructed as follows: ; Taking the actual working conditions and process conditions as constraints, the constraints are constructed and expressed as: In the formula is the minimum value of the i-th influencing factor set according to actual working conditions and process conditions; is the maximum value of the i-th influencing factor set according to actual working conditions and process conditions.

8. The parameter optimization method of the imaging structure of the XRT ore separator according to claim 7, characterized in that The step S5 specifically includes the following steps: The NSGA-Ⅱ algorithm is used to solve the model constructed in step S4 according to the obtained Pareto curve, and the parameter optimization of the imaging structure of the target XRT ore sorter is completed.

9. A system for implementing the parameter optimization method of the XRT ore sorting machine imaging structure according to any one of claims 1 to 8, characterized in that It includes structural parameter acquisition module, simulation model construction module, influencing factor determination module, optimization module construction module and parameter optimization module; The structural parameter acquisition module, the simulation model construction module, the influencing factor determination module, the optimization module construction module and the parameter optimization module are connected in series in sequence; the structural parameter acquisition module is used to obtain structural parameter information of the target XRT ore sorting machine and upload the data information to the simulation model construction module; the simulation model construction module is used to construct a simulation model of the target XRT ore sorting machine in the simulation software based on the received data information and the obtained structural parameter information, in combination with the Monte Carlo simulation scheme, the Boltzmann equation, the cross-validation scheme and the polynomial fitting scheme, and upload the data information to the influencing factor determination module; The influencing factor determination module is used to perform single factor analysis based on the received data information and the obtained simulation model based on the control variable method to study the influence of each element in the X-ray imaging structure on the image signal-to-noise ratio, determine the final influencing factors, and upload the data information to the optimization module construction module; The optimization module construction module is used to construct a parameter optimization model of the imaging structure of the XRT ore sorter based on the received data information and the obtained influencing factors, with the maximum signal-to-noise ratio and the minimum X-ray source power as the optimization goals, and with the actual working conditions and process conditions as constraints, and upload the data information to the parameter optimization module; The parameter optimization module is used to solve the constructed model based on the genetic algorithm according to the received data information, and complete the parameter optimization of the imaging structure of the target XRT ore sorting machine.

Images