Dynamic sorting method and system based on multistage screening and airflow compensation

Abstract

The invention relates to the technical field of screening, in particular to a dynamic sorting method and system based on multistage screening and airflow compensation. The method comprises the steps that 1, space phase screening pretreatment is conducted, specifically, in a rotary type multidirectional screening channel, white rice is made to rotate through a flow guide structure, the included angle between the long axis direction of rice grains and the long edge of a screening hole is obtained in real time, and the rotating angular speed of a screen is dynamically adjusted according to the included angle; 2, mass-airflow coupling sorting is carried out; step 3, performing deformation feedback optical sorting; 4, energy field cooperative control; and 5, dynamic convergence optimization: collecting a screen mesh pore passing rate, a rice grain group distribution entropy value and system total energy consumption power, constructing a state space, and adjusting vibration frequency, airflow pressure and light intensity parameters by adopting a stability judgment algorithm. According to the method, the screening precision is improved, the energy consumption is reduced, the production efficiency is improved, and the method has remarkable practical application value for the grain processing industry.

Description

Technical Field

[0001] The present invention relates to the technical field of screening, and particularly to a dynamic sorting method and system based on multi-stage screening and air flow compensation. Background Art

[0002] In the field of grain processing, especially in the fine processing of white rice, efficient sorting technology is the core link determining the quality of finished products and economic benefits. The existing technologies mainly rely on the method combining mechanical screening and air flow sorting, but there are the following defects in practical applications: Traditional screening equipment adopts a design with fixed-shaped sieve holes and static inclination angles, and cannot adapt to the dynamic change of the aspect ratio of white rice grains (such as the mixed flow of whole grains and broken grains). When there is an angle between the long axis direction of the rice grains and the arrangement direction of the sieve holes, it is easy to cause mis-screening (whole grains are misjudged as broken grains) and missed screening (broken grains are not effectively separated). More seriously, when the rice grain group accumulates on the surface of the sieve mesh, a pore blockage effect occurs, forcing an increase in the vibration intensity, and further causing an increase in the breakage rate of rice grains (the measured broken grain rate generally exceeds 5%); Most existing air flow sorting devices adopt a constant-pressure air supply mode and cannot adjust the air flow parameters according to the real-time changes in the density and falling speed of rice grains. Especially when dealing with rice grains with similar quality but different internal densities (such as worm-eaten rice and intact rice), a single air flow field is difficult to form an effective resistance gradient, resulting in a sharp decline in sorting accuracy (the residual rate of light impurities is as high as 8 - 12%). In addition, the lack of coordinated control of air flow and vibration screening results in low energy utilization efficiency (more than 40% of the system energy consumption is wasted in ineffective turbulence). Summary of the Invention

[0003] Based on the above purposes, the present invention provides a dynamic sorting method and system based on multi-stage screening and air flow compensation. Among them, a dynamic sorting method based on multi-stage screening and air flow compensation includes the following steps: Step 1: Spatial phase screening pretreatment: In a rotary multi-directional screening channel, the white rice is made to form a rotational motion through a diversion structure, the angle between the long axis direction of the rice grains and the long side of the sieve holes is obtained in real time, and the rotational angular velocity of the sieve mesh is dynamically adjusted according to the angle; Step 2: Mass-air flow coupled sorting: A gas-solid two-phase flow mass separation field is constructed at the screening outlet, the density and initial falling velocity of the rice grains are measured, the air flow resistance is calculated, and the elevation angles and air flow velocities of multiple layers of air flow nozzles are configured, and a negative pressure adsorption structure is arranged at the bottom of the separation area; Step 3: Deformation feedback optical sorting: A dual-wavelength laser irradiation system is used to irradiate during the free fall stage of the rice grains and the scattered light intensity distribution is captured, and the rice grains with surface deformation are judged according to the light intensity ratio and sorted through the air flow nozzles; Step 4: Energy field coordinated control: The vibration cycle phase window is divided and the ratio of kinetic energy to potential energy is calculated, and the vibration acceleration and air flow pressure are synchronously adjusted based on the ratio; Step 5: Dynamic Convergence Optimization: Collect the passing rate of the sieve pores, the distribution entropy value of the rice grain group, and the total power consumption of the system, construct a state space, and use a stability determination algorithm to adjust the vibration frequency, air flow pressure, and light intensity parameters.

[0004] Preferably, step 1 includes: The diversion structure is a spiral baffle, and its spiral lift angle is dynamically adjusted according to the average length-to-diameter ratio of the white rice. The surface of the baffle is covered with a material with a low friction coefficient; Continuously capture the movement trajectory of the rice grains through an image acquisition device, extract the main direction of the rice grain contour, and calculate the real-time angle between it and the long side of the sieve hole. The frame rate of the image acquisition device is positively correlated with the flow rate of the rice grains; When the real-time angle exceeds a preset angle threshold, increase the rotational angular velocity of the sieve to above the critical angular velocity, and the critical angular velocity is calculated according to the sieve diameter and the average mass of the rice grains; Set up a vibration detection array on the back of the sieve, collect the vibration signal spectrum generated by the collision of the rice grains, identify the energy ratio of the rigid collision characteristic frequency band and the damping collision characteristic frequency band, and trigger the sieve amplitude compensation mechanism when the damping collision energy ratio exceeds a preset ratio threshold; The amplitude compensation mechanism includes: non-linearly adjusting the longitudinal vibration amplitude of the sieve according to the mapping relationship between the damping collision energy ratio and the sieve porosity, and maintaining the lateral tension of the sieve within the safety threshold range during the adjustment process.

[0005] Preferably, step 2 includes: The multi-layer air flow nozzles are arranged in a multi-layer annular array structure. Each layer contains a plurality of adjustable nozzles evenly distributed circumferentially, and the radial deflection angles of the nozzles between adjacent layers are staggered; When measuring the initial falling velocity of the rice grains, use a non-contact velocity sensing device. Its measurement area covers the entire cross-section of the separation field, and the measurement error caused by the occlusion between the rice grains is eliminated through a time series correlation algorithm; When calculating the air flow resistance, correct the theoretical resistance value according to the standard deviation of the rice grain density distribution, and the correction coefficient has an exponential relationship with the density distribution dispersion; When configuring the nozzle elevation angle, use an iterative approximation algorithm to match the actual air flow velocity field with the theoretical resistance gradient field, and the adjustment amount for each iteration does not exceed a set ratio of the maximum adjustment range of the nozzle; The adsorption strength of the negative pressure adsorption structure is dynamically adjusted according to the impurity concentration detected in real time. The impurity concentration is calculated by measuring the change rate of the transmitted light intensity through an optical sensor array, and the adsorption air flow rate has a non-linear increasing relationship with the impurity concentration increment.

[0006] Preferably, step 3 includes: The dual-wavelength laser irradiation system includes a first-wavelength light source and a second-wavelength light source arranged coaxially, and the wavelength difference between the two is greater than a preset spectral interval threshold; The scattered light intensity distribution is collected using a high-dynamic-range CMOS sensor, whose exposure time is dynamically adjusted according to the light-shielding rate of the falling rice grains, and multi-frame image fusion is performed within each acquisition cycle to eliminate motion blur; When calculating the light intensity ratio, each rice grain projection area is sampled by partitioning, excluding the noise data caused by the edge diffraction effect, and the proportion of the effective sampling area in the total projection area is not less than the set threshold; When determining the surface-deformed rice grains, the sliding window algorithm is used to perform trend analysis on the light intensity ratio data of consecutive rice grains, and when the change rate of the light intensity ratio exceeds the preset fluctuation threshold, a secondary verification process is started; The air nozzle sorting includes multi-stage pressure control. The initial pressure is set according to the median of the rice grain mass distribution, and the subsequent pressure adjustment amount is positively correlated with the deformation determination confidence level.

[0007] Preferably, step 4 includes: The phase window division uses an adaptive time slicing algorithm. The window duration is in a proportional relationship with the reciprocal of the current vibration frequency, and the minimum window duration is limited by the signal response delay of the control system; When calculating the kinetic energy, based on the discrete element simulation model, the motion trajectories of the rice grain group are reconstructed, the instantaneous velocity of each rice grain is denoised by Kalman filtering, and the rice grain data that collides with the sieve is excluded; The potential energy calculation includes the integration of the force exerted by the air pressure field on the rice grain group, and the integration region is dynamically meshed according to the real-time rice grain distribution density; When the ratio of kinetic energy to potential energy exceeds the critical value, a multi-parameter collaborative adjustment strategy is started: the adjustment amount of the vibration acceleration is calculated according to the exceeding amplitude by a piecewise function, and the adjustment amount of the air pressure is negatively feedback associated with the adjustment amount of the vibration acceleration; All parameter adjustment instructions need to pass through the timing verification module to ensure that the control signals of the vibration system and the air flow system are synchronously executed within the preset phase difference range.

[0008] Preferably, step 5 includes: The sieve pore passing rate is calculated through cross-verification of a mass flowmeter and an image recognition system, excluding abnormal data points caused by air flow disturbance; The distribution entropy value of the rice grain group is calculated using an improved Shannon entropy algorithm. The falling trajectory is discretized into three-dimensional grid cells, and the probability density of each cell is determined by the proportion of the residence time of the rice grains in the cell; When constructing the state space, the pore passing rate, the distribution entropy value, and the energy consumption power are dimensionless processed, and dimension reduction is performed to the observable space through the principal component analysis method; The stability determination algorithm includes Lyapunov exponent calculation and phase space reconstruction. When the exponent exceeds the divergence threshold, the multi-objective optimization algorithm is started to synchronously adjust the control parameters; During the parameter adjustment process, the condition number of the system Jacobian matrix is monitored in real time. When the condition number exceeds the preset stability threshold, the current adjustment amount is frozen and the backup control strategy is switched.

[0009] Preferably, the screen amplitude compensation mechanism includes: Establish a transfer function model of the damping collision energy ratio and the amplitude compensation amount, and its non-linear coefficient is calibrated through impact tests; When performing amplitude compensation, a feedforward-feedback composite control structure is adopted: the feedforward control amount is calculated based on the transfer function model, and the feedback control amount is PID-adjusted according to the residual of the damping collision energy ratio after compensation; The transverse tension monitoring is realized by an optical fiber strain sensor embedded in the edge of the screen, and the sampling frequency is higher than a set multiple of the fundamental vibration frequency of the screen; When the transverse tension approaches the safety threshold, an amplitude adjustment cooling period is automatically inserted, during which the vibration energy input is gradually reduced until the tension returns to the safe range.

[0010] Preferably, the iterative approximation algorithm includes: Initialize the nozzle elevation angle to a preset percentage of the theoretical optimal angle, and set the maximum number of iterations and the convergence accuracy threshold; In each iteration, the particle swarm optimization algorithm is used to search for the local optimal solution, and a nozzle mechanical adjustment inertia compensation factor is introduced into the particle velocity update formula; The convergence determination condition is that the change amount of the objective function in consecutive multiple iterations is less than the convergence accuracy threshold, and the air velocity field uniformity index reaches the lowest requirement in the grading standard; When the number of iterations reaches the upper limit and convergence is still not achieved, it is automatically switched to the empirical mode based on historical data, and a system calibration alarm is triggered.

[0011] Preferably, the secondary verification process includes: Perform multi-spectral scanning on the rice grains with an excessive change rate of the light intensity ratio, irradiate with at least two additional auxiliary wavelengths, and reconstruct the three-dimensional model of the surface deformation; Input the feature vector of the three-dimensional model into the pre-trained neural network classifier to output the confidence score of the deformation type; When the confidence score is lower than the determination threshold, the rice grain is marked as an object to be rechecked and temporarily stored in the buffer isolation area; A contact deformation detection device is set in the buffer isolation area, and the micro-force sensor array is used to measure the deformation recovery curve of the rice grains under pressure. The final sorting decision is generated by comprehensively considering the optical determination and contact detection results.

[0012] Correspondingly, an embodiment of the present invention further provides a dynamic sorting system based on multi-stage screening and air flow compensation for operating a dynamic sorting method based on multi-stage screening and air flow compensation according to an embodiment of the present invention, including: A rotary screening pretreatment module, including: A spiral diversion unit configured with a diversion plate having a variable spiral lift angle, the surface of which has a friction coefficient less than the static friction coefficient between white rice and metal; A screen attitude adjustment unit including a rotary screen frame driven by a servo motor, which is rigidly connected to the screen shaft through a coupling; A motion trajectory acquisition unit composed of an array of high-speed industrial cameras, the optical axes of the cameras are installed at an inclined angle to the screen plane, and the output end is signal-connected to the screen attitude adjustment unit; A vibration spectrum analysis unit including a piezoelectric sensor array attached to the back of the screen and a signal conditioning circuit, the output end of which is connected to a screen amplitude compensation controller; A mass-air flow coupling sorting module, including: A velocity field measurement unit composed of an array of laser Doppler velocimeters, and the measurement area covers the entire cross-section of the air flow separation cavity; A multi-layer air flow control unit including coaxially nested multi-layer annular nozzle brackets, and each layer of brackets is evenly distributed with a plurality of adjustable nozzles in the circumferential direction, and the elevation angle of the nozzles is driven by a stepper motor; A negative pressure adsorption unit composed of a centrifugal fan, a cavity pressure sensor and an array of solenoid valves, and an optical impurity detection window is provided at the cavity inlet; A deformation optical sorting module, including: A dual-wavelength light source unit including a first laser emitter and a second laser emitter coaxially installed, and the difference between the two wavelengths is greater than a preset spectral interval; A scattered light acquisition unit composed of a high-frame-rate CMOS sensor and an optical beam splitter, and the output end of the sensor is connected to an image processing board; A pneumatic sorting execution unit including a high-pressure air source, a proportional valve and an array of nozzles, and the outlet direction of the nozzles is arranged orthogonally to the falling trajectory of the rice grains; An energy collaborative control module, including: A vibration phase control unit including an encoder coupled to the screen drive shaft and a vibration exciter, and the output signal of the encoder is connected to a phase segmentation processor; An air flow field strength adjustment unit composed of a pressure transmitter, a proportional integral regulating valve and a flow meter, and its control signal is synchronized with the output of the vibration phase control unit; A dynamic optimization decision module, including: A multi-source data acquisition unit connected to a screen porosity sensor, a distribution entropy calculation chip and a power measurement module; The central processing unit, which has a built-in state space reconstruction algorithm and a core for calculating Lyapunov exponents, is connected to each actuator through a CAN bus at its output end; The connection relationships between the modules are as follows: The outlet of the sieve mesh of the rotary screening pretreatment module is connected to the inlet of the mass-airflow coupling sorting module through an airtight connection pipeline, and the inner wall of the pipeline is provided with an electrostatic elimination coating; The light source unit of the deformation optical sorting module and the scattered light collection unit achieve optical path synchronization through an optical fiber bundle, and the output end of the image processing board is electrically connected to the control terminal of the pneumatic sorting actuator unit; The output end of the vibration phase control unit of the energy co-control module is interconnected with the servo motor driver of the rotary screening pretreatment module by signal, and the pressure feedback signal of the air flow field strength adjustment unit is connected to the nozzle controller of the mass-airflow coupling sorting module; The multi-source data acquisition unit of the dynamic optimization decision module receives the real-time data of each sensor through an industrial Ethernet, and the control instructions generated by the central processing unit are simultaneously sent to the vibration phase control unit, the air flow field strength adjustment unit, and the pneumatic sorting actuator unit.

[0013] Advantages of the present invention: Through the close connection and mutual cooperation of the above-mentioned various steps, the sorting method can achieve efficient and accurate white rice screening. Each step optimizes the sorting effect through a feedback mechanism and co-control, ensuring the high-quality separation of white rice. This method not only improves the screening accuracy, but also reduces energy consumption and improves production efficiency, which has significant practical application value for the grain processing industry. Description of the drawings

[0014] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only those of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0015] Figure 1 It is the step flow chart of the method of the present invention; Figure 2 It is the step flow chart of step 4 of the method of the present invention; Figure 3 It is the structural block diagram of the system of the present invention. Detailed implementation manners

[0016] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. At the same time, it should be noted here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments. For some well-known technologies, those skilled in the art can also adopt other alternative methods for implementation; moreover, the accompanying drawings are only for more specifically describing the embodiments and are not intended to specifically limit the present invention.

[0017] Please refer to Figures 1-3 , an embodiment of the present invention provides a dynamic sorting method based on multi-stage screening and air flow compensation. In step 1, a rotary multi-directional screening channel and a diversion structure are adopted to make the rice grains form a rotary motion. In this way, the cooperation angle between the motion path of the rice grains and the sieve holes is dynamically adjusted. The design of the diversion structure can adjust the rotational angular velocity of the sieve mesh in real time according to the physical characteristics of the rice grains (such as the long axis direction). Through this operation, it is ensured that the rice grains enter the sieve mesh at the best angle, thereby improving the screening accuracy and efficiency.

[0018] This step dynamically adjusts the angular velocity of the sieve mesh, making the cooperation between the rice grains and the sieve holes more precise, ensuring the high efficiency of the preliminary screening, and providing a stable basis for the subsequent sorting.

[0019] In step 2, a gas-solid two-phase flow mass separation field is constructed, and the air flow resistance is calculated by measuring the density and the initial falling velocity of the rice grains. Subsequently, the elevation angle and the air flow velocity of the multi-layer air flow nozzles are configured so that the air flow can effectively separate the rice grains. A negative pressure adsorption structure is arranged at the bottom of the separation area for adsorbing the impurity rice grains, thereby further improving the purity of the sorting.

[0020] Through the coupling of the air flow and the screening, impurities can be effectively removed during the screening process, and the configuration of the air flow can be adjusted according to the physical characteristics of the rice grains, thereby improving the sorting accuracy and ensuring the efficient separation of the rice grains.

[0021] In the free-falling stage of the rice grains in step 3, a dual-wavelength laser irradiation system and the capture of the scattered light intensity distribution are adopted, and whether the rice grains are deformed is judged by analyzing the light intensity ratio. According to the change of the light intensity ratio, the deformed rice grains are sorted through the air flow nozzles. During this process, through the high-precision optical sensor, the accurate detection of the deformed rice grains is ensured.

[0022] This step uses advanced optical feedback technology to accurately identify the subtle deformation on the surface of the rice grains, thereby further improving the screening accuracy. By dynamically adjusting the sorting process of the air flow nozzles, it is ensured that the deformed rice grains can be effectively separated, reducing the screening error.

[0023] In step 4, by dividing the phase window of the vibration period and calculating the ratio of kinetic energy to potential energy, the vibration acceleration and air flow pressure are synchronously adjusted. This method ensures that the vibration and air flow field during the screening process can be adjusted in real time according to the dynamic changes of the rice grains, thereby achieving the optimal sorting effect.

[0024] Through the coordinated control of the energy field, the cooperation between vibration and air flow can be effectively adjusted, the stability of the screening system can be enhanced, and the screening efficiency and the accuracy of rice grain separation can be ensured.

[0025] Finally, in step 5, by collecting data such as the pore passing rate of the sieve mesh, the distribution entropy value of the rice grain population, and the total energy consumption power of the system, a state space is constructed and the vibration frequency, air flow pressure, and light intensity parameters are adjusted through a stability determination algorithm. This process can automatically adjust various parameters during the operation of the entire screening system to ensure the stability and efficiency of the system.

[0026] This step realizes the automatic adjustment of system parameters through a dynamic convergence optimization algorithm, ensuring that the system can continuously maintain the best working state during the screening process regardless of the changing external environment or the change of rice grain types, improving the screening accuracy and the stability of the system.

[0027] Through the close connection and mutual cooperation of the above steps, this sorting method can achieve efficient and accurate white rice screening. Each step optimizes the sorting effect through a feedback mechanism and coordinated control, ensuring the high-quality separation of white rice. This method not only improves the screening accuracy, but also reduces energy consumption and improves production efficiency, which has significant practical application value for the grain processing industry.

[0028] In a possible implementation, first, a spiral deflector is used as the deflector structure. The spiral lift angle is dynamically adjusted according to the average length-diameter ratio of white rice. Through this design, the deflector can more effectively guide the rice grains to move along a predetermined trajectory, avoiding the chaotic distribution caused by the rotation and mutual collision of the rice grains. In addition, the surface of the deflector is covered with a material with a low friction coefficient, reducing the friction force between the rice grains and the deflector, thereby reducing the energy loss during movement and improving the sorting efficiency.

[0029] This design can effectively control the movement trajectory of the rice grains, optimize the screening effect, reduce the damage of the rice grains during the screening process, and at the same time improve the energy utilization efficiency of the entire system.

[0030] Furthermore, the movement trajectory of the rice grains is captured in real time by an image acquisition device. This device adjusts its frame rate according to the change of the rice grain flow rate, so as to more accurately extract the main direction of the rice grain contour and calculate the real-time angle between the rice grain and the long side of the sieve hole. This method uses dynamic image analysis technology to monitor the change of the rice grain movement direction in real time and make timely responses.

[0031] When the real-time angle between the rice grain and the long side of the sieve hole captured by the image acquisition device exceeds the preset angle threshold, the system automatically increases the rotational angular velocity of the sieve to above the critical angular velocity. The critical angular velocity is calculated based on the sieve diameter and the average mass of the rice grains. This operation can ensure that the rice grains enter the sieve at an appropriate speed, avoiding uneven sorting caused by inappropriate angles.

[0032] A vibration detection array is set on the back of the sieve. This array can collect the vibration signal spectrum generated by the collision of rice grains. By analyzing the energy ratio of the rigid collision characteristic frequency band and the damped collision characteristic frequency band in the spectrum, when the energy ratio of the damped collision exceeds the preset ratio threshold, the system will automatically trigger the sieve amplitude compensation mechanism.

[0033] After the amplitude compensation mechanism is triggered, the system non-linearly adjusts the longitudinal vibration amplitude of the sieve according to the mapping relationship between the damped collision energy ratio and the sieve porosity. During this process, it is also necessary to ensure that the lateral tension of the sieve remains within the safe threshold range to avoid damage or deformation of the sieve caused by excessive adjustment.

[0034] Through the detailed design and implementation of the above steps, this method makes full use of the dynamic adjustment and feedback mechanism to ensure that each link can work efficiently and precisely in coordination. The integration of multiple technologies such as the diversion structure, image acquisition, vibration monitoring and compensation mechanism effectively improves the screening accuracy, efficiency and system stability in the white rice sorting process. Each step is closely linked to ensure that the rice grains can be sorted in the best way, thereby improving the overall performance of the sorting system.

[0035] In a possible implementation, first, a multi-layer annular array structure is used to arrange the air nozzles. The nozzles in each layer are evenly distributed circumferentially, and the radial deflection angles of the nozzles between adjacent layers are staggered. This design helps to optimize the distribution of the air flow, enabling the air flow to act on the white rice grains more evenly, thereby improving the accuracy of rice grain sorting. Each nozzle can adjust the direction and speed of the air flow to ensure the best air flow distribution effect.

[0036] By arranging the multi-layer nozzles in a staggered manner, the non-uniformity of the air flow can be effectively reduced, avoiding the concentration of air flow impact caused by a single layer of nozzles, thereby improving the separation effect of white rice grains, especially the adaptability to different characteristics (such as shape, size, density, etc.) of white rice during the sorting process.

[0037] During the measurement of the initial falling velocity of the rice grains, a non-contact velocity sensing device is used. This device can cover the cross-section of the entire separation field and eliminate the measurement error caused by occlusion between rice grains by using the time series correlation algorithm. This design can ensure the accurate measurement of the falling velocity of each rice grain, providing data support for subsequent air flow adjustment.

[0038] The use of a non-contact speed sensor avoids the errors or damage problems caused by direct contact, and the algorithm is used to remove the errors caused by occlusion, which can ensure the accuracy of speed measurement. The accurate initial speed data provides an accurate basis for air flow adjustment, thus effectively improving the efficiency and accuracy of sorting.

[0039] When calculating the air flow resistance, the theoretical resistance value is corrected according to the standard deviation of the rice grain density distribution. The correction coefficient has an exponential relationship with the dispersion degree of the density distribution. The larger the standard deviation of the density distribution, the more significant the influence of the correction coefficient, so that the calculation of the air flow resistance is more accurate and can better meet the sorting requirements of different types of rice grains.

[0040] The calculation method of the corrected air flow resistance enables the system to handle rice grains with different densities and dynamically adjust the air flow resistance according to the actual situation, so as to achieve more refined sorting. By introducing density correction, the mis-sorting of rice grains with different densities is avoided, and the universality and accuracy of the system are enhanced.

[0041] When configuring the nozzle elevation angle, an iterative approximation algorithm is adopted to make the actual air flow velocity field match the theoretical resistance gradient field. The adjustment amount of each iteration is limited within a set ratio of the maximum adjustment range of the nozzle. In this way, the air flow velocity field can be continuously adjusted and optimized, so as to achieve the ideal effect of matching the falling speed and resistance gradient of the rice grains.

[0042] The iterative approximation algorithm can accurately adjust the elevation angle of the nozzle, making the air flow velocity field highly match the motion state of the rice grains, ensuring the high efficiency and accuracy of white rice sorting. The introduction of this algorithm improves the intelligent level of the system and can dynamically adapt to different sorting conditions and requirements.

[0043] The adsorption strength of the negative pressure adsorption structure is dynamically adjusted according to the impurity concentration detected in real time. The impurity concentration is calculated by measuring the change rate of the transmitted light intensity through an optical sensor array. There is a non-linear increasing relationship between the adsorption air flow rate and the impurity concentration increment. When the impurity concentration increases, the adsorption strength will automatically increase to ensure the efficient removal of impurities.

[0044] Dynamically adjusting the adsorption strength can effectively remove impurities and improve the purity of white rice. The optical sensor array provides real-time feedback for the adjustment of the adsorption strength, ensuring that during the sorting process, as the impurity concentration changes, the system can adjust the flow rate of the adsorption air flow in real time to achieve the best sorting effect. This mechanism improves the automation degree and accuracy of the sorting system.

[0045] Through close cooperation and dynamic adjustment, fine control of the white rice sorting process is achieved in aspects such as air nozzle arrangement, speed measurement, air flow resistance correction, nozzle elevation angle adjustment, and impurity removal. Each step is adjusted according to the characteristics of the rice grains and real-time feedback, ensuring high efficiency and accuracy in the sorting process. At the same time, the system can adapt to different working conditions in real time, with strong adaptability and intelligent advantages. Through the organic combination of these technical means, the sorting quality and efficiency of white rice are effectively improved.

[0046] In a possible implementation, the system uses a first wavelength light source and a second wavelength light source arranged coaxially, and the wavelength difference between the two is greater than a preset spectral interval threshold. This design ensures that after the two laser sources penetrate the surface of the white rice, there are significant differences in their reflection and scattering responses, which is beneficial for identifying surface structure details and deformation characteristics. Through coaxial irradiation, the consistency of imaging of the two wavelengths on the rice grains can be guaranteed, avoiding errors caused by differences in incident angles.

[0047] The dual-wavelength scheme improves the recognition sensitivity to minute deformations on the surface of white rice, and is particularly suitable for judging whether the rice grains are damaged, yellowed, or have mechanical cracks.

[0048] The scattered light is collected by a high-dynamic-range CMOS sensor, and its exposure time is dynamically adjusted according to the light-shielding rate of the rice grains, ensuring clear imaging regardless of the large difference in the light transmittance of the rice grains. At the same time, the system performs multi-frame image fusion processing in each acquisition cycle to eliminate motion blur caused by the high-speed falling of the rice grains.

[0049] Dynamic exposure combined with multi-frame fusion significantly improves the imaging clarity, enabling accurate capture of the morphological information of rice grains at different falling speeds and different transparencies, and improving the robustness of the system.

[0050] When calculating the light intensity ratio, the system will perform grid partition sampling on the projection area of each rice grain, automatically excluding abnormal light spots caused by edge diffraction. To ensure data quality, the area of the effective sampling area must account for no less than a preset threshold (such as 85%) of the total projection area of the rice grain to ensure the reliability of the light intensity data for judgment.

[0051] By screening the area to exclude interference information, the light intensity ratio analysis is made more accurate, avoiding misjudgment of the rice grain state due to edge effects and improving the deformation recognition accuracy.

[0052] To judge whether there is deformation in the rice grains, the system uses a sliding window algorithm to perform trend analysis on the light intensity ratio data of continuously passing rice grains. If the change rate of the light intensity ratio in a certain section exceeds the preset fluctuation threshold, it is regarded as possible deformation, and at this time, a secondary verification process is triggered, such as switching to higher-resolution analysis or further judgment in combination with image morphological features.

[0053] Dynamic trend analysis is more adaptable than static threshold judgment. It can identify rice grains with subtle and irregular deformations, while reducing the false alarm rate, making the entire sorting process more intelligent and accurate.

[0054] In the final sorting stage, the air nozzle adopts multi-stage pressure control. The initial pressure is set according to the median of the mass distribution of all detected rice grains to ensure the sorting accuracy at the average level. Subsequently, the injection pressure is dynamically adjusted according to the deformation determination confidence of each rice grain. The higher the confidence, the greater the pressure, to ensure that suspicious deformed rice grains are effectively removed.

[0055] This confidence-linked nozzle pressure control strategy can achieve "applying force as needed", which not only avoids excessive removal of intact rice grains but also ensures that high-confidence defective rice grains are strongly separated, improving the sorting efficiency and the yield rate of good products of the system.

[0056] Through the combination of dual-wavelength illumination and high-dynamic image acquisition, high-quality light intensity data is obtained; then, combined with the region screening and trend analysis algorithms, the rice grains with surface deformations are accurately identified; finally, the sorting process is carried out through an intelligently linked air nozzle, forming an intelligent sorting process with a data closed-loop and adaptive adjustment. This connection method greatly improves the recognition accuracy and sorting efficiency of the surface defects of white rice, and is an important technical means for high-quality white rice production.

[0057] In a possible implementation, the division of the phase window adopts an adaptive time slicing algorithm, which ensures that the duration of the phase window is proportional to the reciprocal of the current vibration frequency of the rice grain. Since the vibration frequencies of different rice grains are different, this proportional relationship enables the system to capture the dynamic characteristics of the rice grains more precisely, thereby improving the sorting accuracy. At the same time, the minimum window duration is limited by the signal response delay of the control system, avoiding information loss or misjudgment caused by too short windows.

[0058] The introduction of the adaptive time slicing algorithm makes the window division more flexible, can respond to the vibration states of different rice grains in real time, and provides a more accurate dynamic characteristic analysis, which helps to better identify the physical characteristics of the rice grains, such as surface deformations or size changes.

[0059] The process of kinetic energy calculation is based on a discrete element simulation model to reconstruct the movement trajectories of the rice grain group during vibration or under the action of air flow. In order to accurately obtain the instantaneous velocity of each rice grain, Kalman filtering is used to denoise the movement data to remove the errors caused by measurement noise or environmental factors. In addition, the data of all rice grains that collide with the sieve mesh will be excluded, so as to ensure that only the influence of the rice grains that do not contact the sieve mesh on the sorting process is considered.

[0060] Through the noise reduction process of Kalman filtering, the system can accurately capture the true motion trajectory of rice grains in an environment with strong noise interference, improve the accuracy of kinetic energy calculation, and thus enhance the stability and reliability of sorting.

[0061] In potential energy calculation, the system integrates the force exerted by the air flow pressure field on the rice grain group, and the integration region will be dynamically meshed according to the real-time distribution density of rice grains. When the rice grain density is high, the mesh refinement is more precise to more accurately simulate the effect of the air flow on the rice grains. This potential energy calculation reflects the interaction between the air flow and the rice grains, affecting the movement trajectory and sorting effect of the rice grains.

[0062] Dynamic mesh refinement can ensure that the force exerted by the air flow pressure field on the rice grain group is accurately simulated, thereby optimizing the regulation of the air flow nozzle and ensuring a more efficient interaction between the air flow and the rice grains during the sorting process.

[0063] When the ratio of kinetic energy to potential energy exceeds the critical value, the system activates the multi-parameter collaborative adjustment strategy. This strategy first calculates the adjustment amount of the vibration acceleration through a piecewise function according to the exceeding amplitude of the ratio of kinetic energy to potential energy. The adjustment amount of the air flow pressure is negatively feedback-correlated with the adjustment amount of the vibration acceleration, that is, when the vibration acceleration increases, the air flow pressure decreases accordingly to avoid uneven sorting caused by over-adjustment of the system.

[0064] Through the collaborative adjustment strategy, the vibration system and the air flow system can cooperate with each other in different dynamic environments, avoiding over-adjustment of a single system. The negative feedback mechanism can ensure the smooth progress of the sorting process, improve the sorting accuracy and avoid over-operation.

[0065] All parameter adjustment instructions must pass through the timing verification module, which ensures that the control signals of the vibration system and the air flow system are synchronously executed within the preset phase difference range. Through precise timing control, the system ensures that the effects of vibration and air flow complement each other within each adjustment cycle, thereby achieving the best sorting effect.

[0066] Timing verification ensures the synchronization of the vibration system and the air flow system, avoids interference caused by the time difference between the two operations, and ensures the stability and sorting accuracy of the system.

[0067] Through adaptive time slicing, precise analysis of kinetic and potential energy calculations, introduction of the collaborative adjustment strategy, and synchronous execution of the timing verification module, intelligent regulation of the dynamic sorting process of white rice is achieved. Through the precise correlation of phase, energy, and control signals between each sub-step, the entire sorting process is optimized. In this way, the system can not only effectively identify and remove non-conforming rice grains, but also ensure the accurate sorting of good-quality rice grains, improving the sorting efficiency, quality, and robustness of the system.

[0068] In a possible implementation, the passing rate of the sieve pores, as a key parameter for evaluating the efficiency of rice grains passing through the sieve, collects real flow data through a mass flowmeter and cross-verifies it with the actual number of rice grains passing through identified by the image recognition system. This cross-mechanism can effectively eliminate abnormal counting points caused by air flow disturbances and improve the accuracy and reliability of the data.

[0069] Avoid control deviations caused by the failure of a single sensor or reading errors, improve the system's ability to reflect the actual physical passing conditions, and provide a high-quality data basis for subsequent judgments.

[0070] Through an improved Shannon entropy algorithm, the system discretizes the trajectory of rice grains during the falling process into three-dimensional grid cells, calculates the proportion of residence time in each cell, and thereby derives the probability density of the cell. The overall entropy value reflects the order degree of the rice grain distribution and the system disturbance state.

[0071] The distribution entropy can be used as an important indicator reflecting the spatial chaos degree of the rice grain group, assisting in judging whether the screening system deviates from the normal sorting path and helping to identify abnormal trends at an early stage.

[0072] Nondimensionalize key variables such as the passing rate of pores, distribution entropy value, and energy consumption power to eliminate the interference of unit dimensions between different physical quantities. Then, through principal component analysis (PCA), the high-dimensional variable space is reduced in dimension and mapped to a low-dimensional observable space for subsequent stability analysis and control judgment.

[0073] Nondimensionalization and PCA dimension reduction improve the calculation efficiency and model generalization ability, simplify the structure of the subsequent judgment model, and at the same time maintain the key dynamic characteristics of the system and improve the control response efficiency.

[0074] The system performs Lyapunov exponent calculation and phase space reconstruction on the state space trajectory. When the Lyapunov exponent exceeds the preset divergence threshold, it means that the system is in an unstable state or tending to a chaotic state. At this time, the multi-objective optimization algorithm is immediately started to comprehensively consider objectives such as efficiency, stability, and energy consumption, and synchronously adjust the control parameters.

[0075] The Lyapunov exponent has high sensitivity and can detect tiny unstable trends of the system in advance, thereby realizing early warning regulation and ensuring that the system maintains a stable operation range for a long time.

[0076] During the adjustment process, the system continuously monitors the condition number of the Jacobian matrix of the control equation, which reflects the system response sensitivity and numerical stability. When the condition number exceeds the preset stability threshold, it indicates that the system may enter an uncontrollable state due to parameter adjustment. The system will freeze the current adjustment action and automatically switch to the backup control strategy to ensure operation safety.

[0077] Introducing the condition number of the Jacobian matrix as a real-time criterion can effectively avoid over-regulation or instability of control signals, and the immediate switching of the backup strategy ensures that the system has good fault self-healing ability and robustness.

[0078] By integrating and analyzing multi-source data (such as screen passing rate, rice grain distribution status, energy consumption power), an observable state space is constructed and real-time stability assessment is implemented, thereby achieving multi-objective collaborative optimization control. Through the mechanism combining Lyapunov criterion and condition number monitoring, not only the adaptive ability of the system to dynamic disturbances is improved, but also a highly intelligent stability guarantee strategy is provided for the sorting system, effectively improving the efficiency, consistency of white rice sorting and the long-term stability of system operation.

[0079] In a possible implementation, first, the system quantifies the energy loss and amplitude relationship of the vibration system under different vibration states by establishing a transfer function model between the damping collision energy ratio and the amplitude compensation amount. The non-linear coefficient of this transfer function needs to be calibrated through impact tests to ensure that the model can accurately reflect the energy loss characteristics in actual operations.

[0080] The transfer function model provides an accurate basis for amplitude compensation calculation, enabling the compensation amount to be adjusted quantitatively according to the actual energy loss, thus achieving more accurate amplitude control in a dynamic environment and avoiding over-compensation or under-compensation.

[0081] In the amplitude compensation process, a feedforward-feedback composite control structure is adopted. Among them, the feedforward control is based on the previously established transfer function model, and the adjustment is made in advance by calculating the theoretically amplitude compensation amount. The feedback control then performs PID adjustment according to the residual of the damping collision energy ratio after compensation to correct the compensation amount in real time. The PID control can automatically adjust the response speed and accuracy of the system, further optimizing the amplitude adjustment effect.

[0082] The composite control structure combines the predictability of feedforward and the real-time adjustment ability of feedback, enabling the system to make rapid and effective adjustments in the face of vibration changes and disturbances, ensuring that the screen amplitude is always maintained within the optimal working range.

[0083] The lateral tension is a key factor affecting the stability of the screen. The system realizes real-time monitoring by embedding fiber optic strain sensors at the edges of the screen. This sensor can collect data on the change of the lateral tension of the screen at a high frequency, and its sampling frequency is set to a certain multiple higher than the fundamental frequency of the screen vibration to ensure that the minute changes during the vibration process can be accurately captured.

[0084] When the system detects that the lateral tension is approaching the safety threshold, to prevent the screen from being damaged due to excessive tension or affecting normal operation, the system will automatically insert an amplitude adjustment cooling period. During this period, the system will gradually reduce the vibration energy input and decrease the vibration amplitude until the lateral tension returns to the safe range. At this time, the cooling period ends, and the system will resume the normal vibration state.

[0085] This screen amplitude compensation mechanism ensures that the screen always maintains a stable and efficient vibration state under different working conditions through precise model calculations, composite control strategies, and real-time tension monitoring. Through the transfer function model calibrated by the nonlinear coefficient, the combination of feedforward and feedback control, as well as the precise monitoring of the lateral tension and the introduction of the cooling period, not only improves the stability and safety of the system, but also can effectively extend the service life of the equipment, reduce the maintenance cost, thus ensuring the efficiency, accuracy, and safety of the white rice sorting process.

[0086] In a possible implementation, the starting step of the algorithm is to initialize the nozzle elevation angle, which is set to a preset percentage of the theoretical optimal angle. This means that in the initial stage, the elevation angle of the nozzle is close to the theoretical value but has not yet reached the optimal state. This step also includes setting the maximum number of iterations and the convergence accuracy threshold. The maximum number of iterations ensures that the algorithm does not fall into an infinite loop, while the convergence accuracy threshold defines the criterion for the algorithm to determine whether the optimal solution has been reached.

[0087] By setting the nozzle angle close to the theoretical optimal value, the number of iterations can be reduced, thereby improving the calculation efficiency and accelerating the convergence process of the system.

[0088] In each iteration, the system uses the Particle Swarm Optimization (PSO) algorithm to search for the local optimal solution and optimize the nozzle elevation angle. The Particle Swarm Optimization algorithm can simulate the search behavior of particle swarms in nature, thus quickly finding a better solution. To further improve the accuracy of the algorithm, a nozzle mechanical adjustment inertia compensation factor is introduced into the particle velocity update formula. The role of this factor is to consider the inertial effect during the nozzle adjustment process, optimize the adjustment action, make the search process of the particles more stable, and avoid over-adjustment or slow adjustment.

[0089] After introducing the inertia compensation factor, the algorithm can better simulate the physical characteristics of the nozzle adjustment, avoid the errors and sluggish response caused by mechanical inertia, and improve the accuracy of the algorithm and the response speed of the system.

[0090] During the iterative process, the algorithm continuously calculates the change in the objective function and determines whether the convergence criterion is met through the convergence determination condition. Specifically, the convergence criterion is that the change in the objective function for consecutive multiple iterations is less than the convergence accuracy threshold, and the uniformity index of the airflow velocity field meets the minimum requirement in the grading standard. This means that during the iterative process, if the change in the objective function tends to be stable and the airflow uniformity has reached the requirement, it is determined that the algorithm has converged and the iteration can be terminated.

[0091] The convergence determination condition ensures that the algorithm can terminate at the appropriate time, avoiding meaningless excessive iterations. It not only optimizes computing resources but also ensures algorithm accuracy, avoiding unnecessary error accumulation.

[0092] If the algorithm does not converge within the preset maximum number of iterations, the system will automatically switch to the empirical mode based on historical data and trigger a system calibration alarm. The empirical mode provides an approximately optimized nozzle adjustment scheme based on past operation data and historical experience. The system calibration alarm alerts the operator that the system has not fully converged and provides an opportunity for manual intervention.

[0093] When the algorithm fails to converge within the maximum number of iterations, the system automatically switches to the empirical mode, avoiding system stagnation or failure. This mechanism ensures that the system can continue to operate even when the algorithm cannot converge, and through the alarm, it prompts the operator to make appropriate manual adjustments or maintenance, thereby improving the robustness and reliability of the system.

[0094] This iterative approximation algorithm can efficiently and accurately optimize the nozzle elevation angle and improve the airflow uniformity by combining particle swarm optimization and the nozzle mechanical adjustment inertia compensation factor, thereby enhancing the effect of the white rice dynamic sorting process. The reasonable settings in the initialization stage, the inertia compensation during the iterative process, the precise convergence determination condition, and the empirical mode switching mechanism in extreme cases ensure that the system can adaptively optimize and operate efficiently and stably under different conditions. Through these steps, the algorithm not only improves the sorting accuracy but also enhances the stability and operating efficiency of the system.

[0095] In a possible implementation, for rice grains with an excessive change rate of light intensity ratio, the system will initiate multi-spectral scanning. This means that the original single-wavelength light source is replaced or supplemented with at least two auxiliary wavelengths. The multi-wavelength light source can more accurately capture the surface characteristics of the rice grains, especially the deformations with different responses to the reflection or absorption characteristics of different wavelengths. In this way, the system can perform a more detailed scan of the rice grain surface, thereby reconstructing a three-dimensional deformation model of the rice grain. This model can accurately reflect the deformation of the rice grain surface at different wavelengths and provide basic data for subsequent deformation analysis.

[0096] Through the reconstructed three-dimensional deformation model, the system extracts the deformation feature vectors of the rice grains and inputs these feature vectors into a pre-trained neural network classifier for analysis. Based on historical data and model training, the neural network can classify the deformation types of the rice grains and output a deformation type confidence score for each rice grain. This score reflects the confidence level of the neural network in determining whether the rice grain conforms to a certain deformation pattern.

[0097] When the deformation type confidence score is lower than the determination threshold, the system deems that the determination of the deformation type of the rice grain is not certain enough. Therefore, it marks the rice grain as an object to be rechecked and temporarily stores it in the buffer isolation area. This process ensures that uncertain rice grains do not directly enter the final sorting decision stage, avoiding possible incorrect sorting.

[0098] In the buffer isolation area, the contact deformation detection device conducts more detailed physical measurements on the rice grains through a micro-force sensor array. The sensor array measures the deformation recovery curve of the rice grains after applying a small pressure, that is, the process of the rice grains returning to their original shape after being pressed. Through this process, the system can further verify the physical properties of the rice grains and determine whether they meet the sorting criteria.

[0099] Finally, the system combines the optical determination result with the contact detection result and makes a final sorting decision based on the judgments of both. This decision integrates the deformation information obtained from optical scanning and the physical feedback of contact deformation detection, ensuring the accuracy and consistency of sorting.

[0100] The secondary verification process realizes the comprehensive detection and verification of rice grain deformation through a series of refined steps such as multi-spectral scanning, neural network classification, and micro-force sensor detection. By temporarily storing and rechecking low-confidence rice grains and combining high-precision physical detection methods, the system can significantly improve the sorting accuracy and ensure high-quality output in the rice grain sorting process. This process not only enhances the robustness of the sorting system but also improves the ability to handle complex deformations, making the application effect of this method in dynamic sorting more ideal.

[0101] Correspondingly, the embodiment of the present invention further provides a dynamic sorting system based on multi-stage screening and air flow compensation for running the dynamic sorting method based on multi-stage screening and air flow compensation described in the embodiment of the present invention, including: A rotary screening preprocessing module, including: A spiral diversion unit, configured with a diversion plate having a variable spiral lift angle, and the surface coating has a friction coefficient less than the static friction coefficient between white rice and metal; A screen attitude adjustment unit, including a rotary screen frame driven by a servo motor, which is rigidly connected to the screen shaft through a coupling; The motion trajectory acquisition unit is composed of an array of high-speed industrial cameras. The optical axes of the cameras are installed at an inclined angle with respect to the screen plane, and the output end is signal-connected to the screen attitude adjustment unit; The vibration spectrum analysis unit includes a piezoelectric sensor array attached to the back of the screen and a signal conditioning circuit, and its output end is connected to the screen amplitude compensation controller; The mass-airflow coupling sorting module includes: The velocity field measurement unit is composed of an array of laser Doppler velocimeters, and the measurement area covers the entire cross-section of the air separation cavity; The multi-layer air flow control unit includes a multi-layer annular nozzle bracket nested coaxially. Each layer of the bracket is evenly distributed with a plurality of adjustable nozzles in the circumferential direction, and the nozzle elevation angle is driven by a stepper motor; The negative pressure adsorption unit is composed of a centrifugal fan, a cavity pressure sensor and an electromagnetic valve array. An optical impurity detection window is provided at the cavity inlet; The deformation optical sorting module includes: The dual-wavelength light source unit includes a first laser transmitter and a second laser transmitter installed coaxially, and the difference between the two wavelengths is greater than the preset spectral interval; The scattered light acquisition unit is composed of a high-frame-rate CMOS sensor and an optical beam splitter, and the output end of the sensor is connected to the image processing board; The pneumatic sorting execution unit includes a high-pressure air source, a proportional valve and an array of nozzles. The outlet direction of the nozzles is arranged orthogonally to the falling trajectory of the rice grains; The energy collaborative control module includes: The vibration phase control unit includes an encoder coupled to the screen drive shaft and a vibration exciter, and the output signal of the encoder is connected to the phase segmentation processor; The air flow field strength adjustment unit is composed of a pressure transmitter, a proportional-integral regulating valve and a flow meter, and its control signal is synchronized with the output of the vibration phase control unit; The dynamic optimization decision-making module includes: The multi-source data acquisition unit is connected to the screen porosity sensor, the distribution entropy calculation chip and the power measurement module; The central processing unit is built-in with a state space reconstruction algorithm and a Lyapunov exponent calculation core, and its output end is connected to each actuator through a CAN bus; The connection relationship between the modules is as follows: The outlet of the screen of the rotary screening pretreatment module is connected to the inlet of the mass-airflow coupling sorting module through an airtight connection pipeline, and the inner wall of the pipeline is provided with an electrostatic elimination coating; The light source unit of the deformation optical sorting module and the scattered light acquisition unit achieve optical path synchronization through an optical fiber bundle, and the output end of the image processing board is electrically connected to the control terminal of the pneumatic sorting execution unit; The output end of the vibration phase control unit of the energy collaborative control module is interconnected with the servo motor driver of the rotary screening preprocessing module, and the pressure feedback signal of the air flow field strength adjustment unit is connected to the nozzle controller of the mass-airflow coupling sorting module; The multi-source data acquisition unit of the dynamic optimization decision-making module receives the real-time data of each sensor through the industrial Ethernet, and the control instructions generated by the central processing unit are simultaneously sent to the vibration phase control unit, the air flow field strength adjustment unit, and the pneumatic sorting execution unit.

[0102] This invention covers any substitutions, modifications, equivalent methods, and solutions made within the essence and scope of this invention. To enable the public to have a thorough understanding of this invention, specific details are elaborated in the following preferred embodiments of this invention. However, those skilled in the art can fully understand this invention without the description of these details. Additionally, to avoid unnecessary confusion to the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0103] The above are only the preferred embodiments of this invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of this invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of this invention.

Claims

1. A dynamic sorting method based on multi-stage screening and airflow compensation, characterized in that: The following steps are involved: Step 1: spatial phase screening pretreatment: in the rotary multi-directional screening channel, the white rice is made to rotate through the guide structure, the angle between the long axis of the rice grain and the long side of the sieve hole is obtained in real time, and the angular velocity of the sieve rotation is dynamically adjusted according to the angle; Step 2: Mass-airflow coupling separation: construct a gas-solid two-phase flow mass separation field at the screening outlet, measure the density and initial falling velocity of rice grains, calculate the airflow resistance and configure the elevation angle and airflow velocity of the multi-layer airflow nozzle, and set a negative pressure adsorption structure at the bottom of the separation area; Step 3: Deformation feedback optical sorting: A dual-wavelength laser irradiation system is used to capture the scattered light intensity distribution during the free-falling stage of rice grains. The surface deformed rice grains are determined based on the light intensity ratio and sorted through an airflow nozzle; Step 4: Energy field coordinated control: Divide the vibration period phase window and calculate the ratio of kinetic energy to potential energy, and synchronously adjust the vibration acceleration and airflow pressure based on the ratio; Step 5: Dynamic convergence optimization: Collect the sieve mesh pass rate, rice grain group distribution entropy value and system total energy consumption power, construct the state space and use the stability judgment algorithm to adjust the vibration frequency, airflow pressure and light intensity parameters.

2. A dynamic sorting method based on multi-stage screening and airflow compensation according to claim 1, characterized in that: The step 1 comprises: The guide structure is a spiral guide plate, the spiral angle of which is dynamically adjusted according to the average aspect ratio of white rice, and the surface of the guide plate is covered with a low friction coefficient material; The movement trajectory of rice grains is continuously captured by an image acquisition device, the main direction of the rice grain contour is extracted, and the real-time angle between the main direction and the long side of the sieve hole is calculated, wherein the frame rate of the image acquisition device is positively correlated with the flow rate of the rice grains; When the real-time angle exceeds a preset angle threshold, the angular velocity of the screen rotation is increased to above a critical angular velocity, wherein the critical angular velocity is calculated based on the screen diameter and the average mass of the rice grains; A vibration detection array is set on the back of the screen to collect the vibration signal spectrum generated by the collision of rice grains, identify the energy ratio of the characteristic frequency band of rigid collision and the characteristic frequency band of damping collision, and trigger the screen amplitude compensation mechanism when the damping collision energy ratio exceeds the preset ratio threshold; The amplitude compensation mechanism includes: nonlinearly adjusting the longitudinal vibration amplitude of the screen according to the mapping relationship between the damping collision energy ratio and the screen porosity, and maintaining the lateral tension of the screen within a safe threshold range during the adjustment process.

3. A dynamic sorting method based on multi-stage screening and airflow compensation according to claim 2, characterized in that: The step 2 comprises: The multi-layer airflow nozzles are arranged in a multi-layer annular array structure, each layer includes a plurality of circumferentially evenly distributed adjustable nozzles, and the radial deflection angles of the nozzles between adjacent layers form a staggered distribution; When measuring the initial velocity of the falling rice grains, a non-contact velocity sensor is used, whose measurement area covers the entire separation field cross section, and the measurement error caused by the occlusion between rice grains is eliminated by the time series correlation algorithm; When calculating the airflow resistance, the theoretical resistance value is corrected according to the standard deviation of the rice grain density distribution, and the correction coefficient is exponentially related to the dispersion of the density distribution; When configuring the nozzle elevation angle, an iterative approximation algorithm is used to match the actual airflow velocity field with the theoretical resistance gradient field, and the adjustment amount of each iteration does not exceed the set proportion of the maximum adjustment range of the nozzle; The adsorption intensity of the negative pressure adsorption structure is dynamically adjusted according to the real-time detected impurity concentration. The impurity concentration is calculated by measuring the change rate of the transmitted light intensity through the optical sensor array. The adsorption airflow flow rate and the impurity concentration increment are in a nonlinear increasing relationship.

4. A dynamic sorting method based on multi-stage screening and airflow compensation according to claim 3, characterized in that: The step 3 comprises: The dual-wavelength laser irradiation system comprises a first wavelength light source and a second wavelength light source which are coaxially arranged, and the difference between the two wavelengths is greater than a preset spectral interval threshold; The scattered light intensity distribution is collected using a high dynamic range CMOS sensor, whose exposure time is dynamically adjusted according to the shading rate of the falling rice grains, and multi-frame image fusion is performed in each acquisition cycle to eliminate motion blur; When calculating the light intensity ratio, the projection area of ​​each rice grain is sampled in different areas to exclude the noise data caused by the edge diffraction effect. The ratio of the effective sampling area to the total projection area is not less than the set threshold. When judging the surface deformed rice grains, a sliding window algorithm is used to perform trend analysis on the light intensity ratio data of consecutive rice grains, and a secondary verification process is initiated when the light intensity ratio change rate exceeds a preset fluctuation threshold; The airflow nozzle sorting includes multi-stage pressure control, the initial pressure is set according to the median of the rice grain mass distribution, and the subsequent pressure adjustment amount is positively correlated with the confidence of deformation judgment.

5. A dynamic sorting method based on multi-stage screening and airflow compensation according to claim 4, characterized in that: The step 4 comprises: The phase window division adopts an adaptive time slicing algorithm, the window duration is proportional to the inverse of the current vibration frequency, and the minimum window duration is limited by the signal response delay of the control system; When calculating kinetic energy, the motion trajectory of the rice grain group is reconstructed based on the discrete element simulation model, the instantaneous velocity of each rice grain is subjected to Kalman filtering and noise reduction, and the data of rice grains that collide with the screen are excluded; The potential energy calculation includes the integration of the force exerted by the airflow pressure field on the rice grain group, and the integration area is dynamically meshed according to the real-time rice grain distribution density; When the ratio of kinetic energy to potential energy exceeds the critical value, the multi-parameter coordinated adjustment strategy is activated: the vibration acceleration adjustment amount is calculated according to the excess amplitude of the ratio using a piecewise function, and the airflow pressure adjustment amount forms a negative feedback relationship with the vibration acceleration adjustment amount; All parameter adjustment instructions must pass the timing verification module to ensure that the control signals of the vibration system and the airflow system are executed synchronously within the preset phase difference range.

6. A dynamic sorting method based on multi-stage screening and airflow compensation according to claim 5, characterized in that: The step 5 comprises: The sieve mesh pore pass rate is calculated by cross-validation of a mass flow meter and an image recognition system, and abnormal data points caused by airflow disturbance are eliminated; The entropy value of the rice grain group distribution is calculated using the improved Shannon entropy algorithm, which discretizes the falling trajectory into three-dimensional grid cells. The probability density of each cell is determined by the proportion of the rice grain residence time in the cell. When constructing the state space, the pore permeability, distribution entropy and energy consumption power are dimensionless and reduced to observable space through principal component analysis. The stability determination algorithm includes Lyapunov exponent calculation and phase space reconstruction. When the exponent exceeds the divergence threshold, the multi-objective optimization algorithm is started to synchronously adjust the control parameters. During the parameter adjustment process, the condition number of the system Jacobian matrix is ​​monitored in real time. When the condition number exceeds the preset stability threshold, the current adjustment amount is frozen and switched to the backup control strategy.

7. A dynamic sorting method based on multi-stage screening and airflow compensation according to claim 6, characterized in that: The screen amplitude compensation mechanism includes: A transfer function model of damping collision energy ratio and amplitude compensation is established, and its nonlinear coefficient is calibrated through impact test. When performing amplitude compensation, a feedforward-feedback composite control structure is used: the feedforward control quantity is calculated based on the transfer function model, and the feedback control quantity is PID-adjusted according to the residual of the damping collision energy ratio after compensation; Transverse tension monitoring is achieved through optical fiber strain sensors embedded in the edge of the screen, with a sampling frequency higher than the set multiple of the fundamental frequency of the screen vibration; When the lateral tension approaches the safety threshold, an amplitude adjustment cooling cycle is automatically inserted, during which the vibration energy input is gradually reduced until the tension returns to a safe range.

8. The dynamic sorting method based on multi-stage screening and airflow compensation according to claim 3 is characterized in that: The iterative approximation algorithm includes: Initialize the nozzle elevation angle to a preset percentage of the theoretical optimal angle, and set the maximum number of iterations and the convergence accuracy threshold; In each iteration, the local optimal solution is searched by particle swarm optimization algorithm, and the nozzle mechanical adjustment inertia compensation factor is introduced into the particle velocity update formula; The convergence judgment condition is that the change of the objective function of multiple consecutive iterations is less than the convergence accuracy threshold, and the airflow velocity field uniformity index meets the minimum requirement of the classification standard; When the number of iterations reaches the upper limit and still fails to converge, it automatically switches to the experience mode based on historical data and triggers the system calibration alarm.

9. A dynamic sorting method based on multi-stage screening and airflow compensation according to claim 4, characterized in that: The secondary verification process includes: Start multi-spectral scanning for rice grains with excessive light intensity ratio change rate, add at least two auxiliary wavelengths for illumination and reconstruct a three-dimensional model of surface deformation; Input the 3D model feature vector into the pre-trained neural network classifier and output the deformation type confidence score; When the confidence score is lower than the judgment threshold, the rice grain is marked as an object to be reviewed and temporarily stored in the buffer isolation area; A contact deformation detection device is set up in the buffer isolation area, and the compression deformation recovery curve of rice grains is measured by a micro-force sensor array. The final sorting decision is generated by integrating optical judgment and contact detection results.

10. A dynamic sorting system based on multi-stage screening and airflow compensation, used to run a dynamic sorting method based on multi-stage screening and airflow compensation according to any one of claims 1 to 9, characterized in that: include: Rotary screening pre-treatment module, including: The spiral guide unit is equipped with a guide plate with a variable spiral angle, and the friction coefficient of the surface coating is less than the static friction coefficient of white rice and metal; A screen posture adjustment unit, comprising a rotating screen frame driven by a servo motor, rigidly connected to the screen shaft through a coupling; The motion trajectory acquisition unit is composed of a high-speed industrial camera array. The camera optical axis is installed at an inclined angle to the screen plane, and the output end is connected to the screen posture adjustment unit signal; A vibration spectrum analysis unit, comprising a piezoelectric sensor array and a signal conditioning circuit attached to the back of the screen, the output end of which is connected to the screen amplitude compensation controller; Mass-airflow coupled sorting module, including: The velocity field measurement unit is composed of a laser Doppler velocimeter array, and the measurement area covers the entire cross-section of the airflow separation cavity; The multi-layer airflow control unit includes coaxially nested multi-layer annular nozzle brackets, each layer of the bracket has multiple adjustable nozzles evenly distributed circumferentially, and the nozzle elevation angle is driven by a stepper motor; The negative pressure adsorption unit is composed of a centrifugal fan, a cavity pressure sensor and a solenoid valve array, and an optical impurity detection window is provided at the cavity entrance; Deformation optical sorting module, including: A dual-wavelength light source unit, comprising a first laser emitter and a second laser emitter mounted coaxially, the difference between the two wavelengths being greater than a preset spectral interval; The scattered light collection unit consists of a high frame rate CMOS sensor and an optical beam splitter, and the sensor output is connected to the image processing board; A pneumatic sorting execution unit includes a high-pressure air source, a proportional valve and an array nozzle, wherein the nozzle outlet direction is arranged orthogonally to the rice grain falling trajectory; Energy collaborative control module, including: A vibration phase control unit, comprising an encoder coupled to the screen drive shaft and a vibration exciter, wherein the encoder output signal is connected to a phase division processor; The airflow field intensive regulation unit is composed of a pressure transmitter, a proportional integral control valve and a flow meter, and its control signal is synchronized with the output of the vibration phase control unit; Dynamic optimization decision module, including: Multi-source data acquisition unit, connecting the screen porosity sensor, distribution entropy calculation chip and power metering module; The central processing unit has a built-in state space reconstruction algorithm and Lyapunov index calculation core, and its output is connected to each actuator through the CAN bus; The connection relationship between modules is as follows: The outlet of the rotary screening pretreatment module is connected to the inlet of the mass-airflow coupling sorting module through an airtight pipe, and the inner wall of the pipe is provided with a static elimination coating; The light source unit of the deformation optical sorting module and the scattered light collection unit realize optical path synchronization through the optical fiber bundle, and the output end of the image processing board is electrically connected to the control terminal of the pneumatic sorting execution unit; The output end of the vibration phase control unit of the energy coordination control module is interconnected with the servo motor driver signal of the rotary screening preprocessing module, and the pressure feedback signal of the airflow field stress regulation unit is connected to the nozzle controller of the mass-airflow coupling sorting module; The multi-source data acquisition unit of the dynamic optimization decision module receives real-time data from each sensor via industrial Ethernet, and the control instructions generated by the central processing unit are simultaneously sent to the vibration phase control unit, the airflow field intensity regulation unit and the pneumatic sorting execution unit.

Images