Cross screen screening parameter optimization method and system based on IGWO-BP-PID

Abstract

The invention provides a cross sieve screening parameter optimization method and a cross sieve screening parameter optimization system based on an (IGWO) algorithm, a back propagation neural network (Back-Propagation Neural Network) and proportional integral derivative (PID) control, and relates to a cross sieve screening parameter optimization method and a cross sieve screening parameter optimization system based on the IGWO algorithm, the back propagation Neural Network (Back-Propagation Neural Derivative (PID). The system comprises a cross sieve, a control system, a communication unit, a monitoring system and an upper computer, the monitoring system monitors whether the screening parameters of the cross screen meet the optimal screening condition or not; when the optimal screening condition is not met, the upper computer optimizes the screening parameters through a preset IGWO-BP-PID algorithm and transmits the optimized screening parameters to a control system; and the control system adjusts the action of a mechanical system in the cross screen according to the optimized screening parameters to enable the mechanical system to meet the optimal screening condition. By means of the system, it can be guaranteed that under different working conditions, a mechanical system in the cross screen can be automatically and quantitatively adjusted, different screen face inclination angles are formed, smooth screening is guaranteed, the system can run in a full-automatic mode, and the screening efficiency is further improved.

Description

Technical Field

[0001] The invention relates to the technical field of screening equipment, and in particular to a cross-screen screening parameter optimization method and system based on IGWO-BP-PID. Background Art

[0002] The cross-type fine-grain roller screen is a screening device suitable for deep dry screening of wet fine raw coal. It has broad application prospects in thermal power plants, coal mines, coal preparation plants and other industries. Although the existing cross screen can provide basic screening functions, it lacks intelligent automatic adjustment capabilities and cannot automatically optimize the screen surface angle according to the physical properties of different materials to make the material evenly distributed. Since the screen surface structure is fixed, it cannot be adaptively adjusted when the working conditions change, resulting in low screening efficiency and uneven screen surface wear. In addition, the equipment lacks material monitoring and real-time adjustment mechanisms, and cannot respond quickly when material properties change, affecting the screening effect and equipment life. Therefore, improving the intelligence level of the equipment, especially the introduction of automatic adjustment and real-time monitoring systems, is the key to solving the current problems. Summary of the invention

[0003] The purpose of the present invention is to provide a cross-screen screening parameter optimization method and system based on IGWO-BP-PID, so that the cross-type fine-grain roller screen can be automatically adjusted to different screen surface inclination angles according to different complex working conditions, thereby facilitating the smooth progress of screening operations.

[0004] In order to solve the above technical problems, an embodiment of the present invention provides a cross-screen screening parameter optimization system based on IGWO-BP-PID, including a cross-type fine-grain roller screen, a control system, a communication unit, a monitoring system, and a host computer; the monitoring system is connected to the host computer through the communication unit, and the monitoring system is used to monitor whether the screening parameters of the cross-type fine-grain roller screen meet the optimal screening conditions; when the optimal screening conditions are not met, the monitoring system transmits the screening parameters to the host computer through the communication unit; the host computer is also connected to the control system, and the host computer reads the screening parameters and optimizes them through a preset IGWO-BP-PID algorithm to obtain optimized screening parameters, and transmits the optimized screening parameters to the control system; the control system is also connected to the cross-type fine-grain roller screen, and the control system is used to adjust the mechanical system action in the cross-type fine-grain roller screen according to the optimized screening parameters so that it meets the optimal screening conditions.

[0005] An embodiment of the present invention also provides a cross-screening parameter optimization method based on IGWO-BP-PID, which is applied to a host computer, including: receiving screening parameters that do not meet the optimal screening conditions sent by a monitoring system; optimizing the screening parameters by a preset IGWO-BP-PID algorithm to obtain optimized screening parameters; and transmitting the optimized screening parameters to a control system.

[0006] In addition, the screening parameters are optimized by the preset IGWO-BP-PID algorithm to obtain the optimized screening parameters, including: establishing a corresponding motion model according to the motion analysis of the cross-type fine-grain roller screen; designing a PID controller according to the system model; initializing the BP neural network, and predicting the control output based on the historical data of the mechanical system of the cross-type fine-grain roller screen; using the IGWO algorithm to generate an initial PID parameter population, each group of parameters represents Kp, Ki, and Kd of the PID controller, where Kp, Ki, and Kd are the proportional, integral, and differential coefficients of the PID controller, respectively; using the initialized BP neural network to calculate the fitness value of each group of PID parameters in the initial PID parameter population; updating the population position based on the fitness value, and outputting the optimal PID parameter combination as the optimized screening parameters to control the mechanical system.

[0007] In addition, using the IGWO algorithm to generate the initial PID parameter population includes: using a chaotic map to generate the initial PID parameter population.

[0008] In addition, in the mechanical system, the PID parameter combination adjusts the extension of the hydraulic cylinder through the following formula to achieve control of the mechanical system;

[0009] Adjusting the extension x(t) of the hydraulic cylinder involves making adjustments using the following formula:

[0010]

[0011] Where u(t) represents the PID output, x(t) is the actual extension of the hydraulic cylinder, and x setpoint represents the target elongation of the hydraulic cylinder, e(t) represents the difference between the actual elongation and the target elongation, t is the time variable representing the elongation process, e(t) = x setpoint -x(t),K p , K i , K d are the proportional, integral, and differential coefficients of the PID controller, respectively, and e(τ) represents the value of the cumulative elongation error at different time points;

[0012] The fitness value is defined based on the performance indicators of the hydraulic cylinder:

[0013] Fitness=w1×ITAE+w2×MSE+w3×Overshoot

[0014]

[0015] Among them, the integral time absolute error ITAE is a measure of the error and response time of the hydraulic cylinder, the mean square error MSE is the steady-state error; Overshoot is the overshoot, which is used to indicate the maximum deviation of the elongation of the hydraulic cylinder before reaching the set value; w1, w2, and w3 are weight coefficients, w1 corresponds to ITAE, w2 corresponds to MSE, and w3 corresponds to Overshoot; T corresponds to the total number of time periods or the total number of sampling points for calculating the error.

[0016] In addition, the motion analysis of the cross-type fine-grain roller screen includes the establishment of the following motion model:

[0017]

[0018] Among them, θ1 is the initial pitch angle, θ2 is the pitch adjustment angle, θ1+θ2 is the angle after pitch control, β1 is the initial swing angle, β2 is the swing adjustment angle, β1+β2 is the angle after swing control, x is the extension length of the hydraulic cylinder in the swing control device, y is the extension length of the hydraulic cylinder in the pitch control device, A is the distance between the rotation center of the inner and outer frame connecting device and the same plane as the pitch hydraulic cylinder and the top plate hinge point, B is the distance between the rotation center of the inner and outer frame connecting device and the pitch hydraulic cylinder connecting seat and the inner frame hinge point, C is the initial length of the pitch hydraulic cylinder, a is the distance from the front support frame hinge point to the support swing hydraulic cylinder and the ground contact support frame hinge point, b is the distance from the front support frame hinge point to the swing hydraulic cylinder and the swing hydraulic cylinder connecting seat hinge point, c is the initial height of the swing hydraulic cylinder.

[0019] In addition, monitoring whether the screening parameters of the cross-type fine-grain roller screen meet the optimal screening conditions includes: establishing a mathematical model of the stockpile; obtaining the screening parameters of the cross-type fine-grain roller screen from the mathematical model and determining whether the optimal screening conditions are met.

[0020] In addition, a mathematical model of the stockpile is established, including:

[0021] Scan the pile of materials through a linear array sensor to obtain point cloud data of the pile of materials;

[0022] Remove noise points from point cloud data and perform point cloud filtering;

[0023] The filtering method is statistical filtering, including calculating the distance of each point to its neighboring points to determine whether the point is a noise point. i =(x i ,y i ,z i ), the average distance d of its neighboring points m for:

[0024]

[0025] Where k is the number of neighboring points, d(p i ,p j ) indicates p i With neighboring point p j distance;

[0026] Then, the surface model of the point cloud data is fitted using the least square method to realize the morphological recognition of the stockpile;

[0027] Specifically, when the least squares method is used to fit the surface model, a quadratic surface equation is used to approximate the point cloud data, and when the pile is scanned by a linear array sensor, three-dimensional point cloud data (x i ,y i ,z i ), and the quadratic surface equation z=ax is obtained by fitting 2 +by 2 +cxy+dx+ey+f, where a, b, c, d, e, and f are fitting coefficients; the specific steps for determining the fitting coefficients are:

[0028] Define the error function:

[0029] Construct the sum of squared errors: Where N is the total number of points,

[0030] Construct a linear equation system: Z = Xβ, where

[0031]

[0032] The fitting parameter β is obtained by solving the following normal equation:

[0033] β=(X T X) -1 X T Z, where X T is the transpose of the matrix X, (X T X) -1 Yes X T The inverse matrix of X;

[0034] The fitting coefficients are extracted to obtain the mathematical model of the surface model.

[0035] In addition, the mechanical system actions for adjusting the cross-type fine-grain roller screen include:

[0036] Scan the pile of materials through a linear array sensor to obtain point cloud data of the pile for swing control;

[0037] The material thickness at the inlet is obtained by the ultrasonic distance sensor for pitch control; the material thickness is calculated by the time difference between the ultrasonic distance sensor emitting a short ultrasonic pulse and the ultrasonic distance sensor receiving the ultrasonic signal reflected from the screen material pile. The specific calculation formula is:

[0038]

[0039] v=331.4+0.6×F

[0040] Wherein, S is the distance between the ultrasonic ranging sensor and the pile, v is the speed of sound, the time difference is K, and the real-time temperature is F.

[0041] The system in the embodiment of the present invention includes a cross-type fine-grain roller screen, a control system, a communication unit, a monitoring system, and a host computer; the monitoring system monitors whether the screening parameters of the cross-type fine-grain roller screen meet the optimal screening conditions; when the optimal screening conditions are not met, the host computer reads the screening parameters and optimizes them through the preset IGWO-BP-PID algorithm to obtain the optimized screening parameters, and transmits the optimized screening parameters to the control system; the control system adjusts the mechanical system action in the cross-type fine-grain roller screen according to the optimized screening parameters to make it meet the optimal screening conditions. The above system can ensure that under different complex working conditions, including different types of screening materials, different stockpiling conditions, etc., the mechanical system in the cross-type fine-grain roller screen can be automatically and quantitatively adjusted to form different screen surface inclination angles, which can be beneficial to cope with different screening operations and ensure smooth screening. The system can also run fully automatically, further improving the screening efficiency.

[0042] Compared with the prior art, the present invention has the following significant effects:

[0043] 1. Through the integration of monitoring, control, optimization and other modules, the system can realize the automatic operation of the screening process, thereby reducing manual intervention and improving screening efficiency. The system can automatically adjust parameters such as the screen surface inclination and roller speed according to different working conditions, adapt to changes in conditions such as material stacking methods, and achieve efficient screening. The system also has a real-time monitoring and feedback mechanism, which can quickly adjust operating parameters when screening conditions change, ensuring that the screening process continues to operate under optimal conditions.

[0044] 2. Through the combination of IGWO and BP neural network, the system can achieve parameter optimization under complex screening conditions, avoiding the problem that traditional optimization algorithms are prone to fall into local optimality in multi-dimensional space. The relationships between many parameters in the screening process are complex and nonlinear. The BP neural network can capture these relationships, optimize screening conditions, and improve screening quality. The PID controller can make timely adjustments to rapidly changing working conditions to prevent delayed adjustments or excessive adjustments, thereby maintaining the stability of the screening process. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a structural schematic diagram of a cross-screen screening parameter optimization system based on IGWO-BP-PID according to one embodiment of the present invention;

[0046] Figure 2 This is a workflow diagram of a cross-screen screening parameter optimization system based on IGWO-BP-PID according to an embodiment of the present invention;

[0047] Figure 3 A front view of a cross-type fine particle roller screen according to an embodiment of the present invention;

[0048] Figure 4 for Figure 3 Right view of . DETAILED DESCRIPTION

[0049] To make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. However, it will be appreciated by those skilled in the art that in the embodiments of the present invention, many technical details are proposed in order to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present application can be implemented. The division of the following embodiments is for the convenience of description, and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined and referenced with each other without contradiction.

[0050] One embodiment of the present invention relates to a method and system for optimizing screening parameters of a cross-type fine-grain roller screen based on an Improved GreyWolf Optimization (IGWO) algorithm, a back-propagation neural network (Back-Propagation Neural Network) and a proportional integral derivative (PID) control. The system can be applied to an integrated screening tool including a cross-type fine-grain roller screen, such as Figure 1As shown, the method can be applied in a host computer. The system in the embodiment of the present invention includes a cross-type fine-grain roller screen, a control system, a communication unit, a monitoring system, and a host computer; the monitoring system is connected to the host computer through the communication unit, and the monitoring system is used to monitor whether the screening parameters of the cross-type fine-grain roller screen meet the optimal screening conditions; when the optimal screening conditions are not met, the monitoring system transmits the screening parameters to the host computer through the communication unit; the host computer is also connected to the control system, and the host computer reads the screening parameters and optimizes them through the preset IGWO-BP-PID algorithm to obtain the optimized screening parameters, and transmits the optimized screening parameters to the control system; the control system is also connected to the cross-type fine-grain roller screen, and the control system is used to adjust the mechanical system action in the cross-type fine-grain roller screen according to the optimized screening parameters to meet the optimal screening conditions. Through the above system, it can be ensured that under different complex working conditions, including different types of screening materials, different stockpiling conditions, etc., the mechanical system in the cross-type fine-grain roller screen can be automatically and quantitatively adjusted to form different screen surface inclination angles, which can be beneficial to cope with different screening operations and ensure smooth screening. The following is a specific description of the optimization of screening parameters and system implementation details of the cross-type fine-grained roller screen based on IGWO-BP-PID according to one embodiment of the present invention. The following content is only the implementation details provided for the convenience of understanding and is not necessary for the implementation of this solution.

[0051] Specifically, the mechanical system of the cross-type fine-grain roller screen includes a cross-type fine-grain roller screen body, an outer frame, an inner frame, a pitch control device, and a swing control device. The screen body is installed on the inner frame, and the inner frame is arranged in the outer frame in a nested manner. The pitch control device is connected to the inner frame and the outer frame. The screen body can adjust the pitch angle relative to the inner screen frame through the pitch control device. The swing control device is connected to the ground and the outer frame, wherein the inner frame and the outer frame are connected together, and the screen machine as a whole can be swung through the pitch control device.

[0052] The monitoring system includes linear array laser sensors, ultrasonic distance sensors, inclination sensors, displacement sensors, and rotary encoders. The linear array laser sensor is installed near the roller screen outlet, so that the material distribution on the entire screen surface can be scanned; since the roller screen is most likely to have excessive material thickness at the inlet, the ultrasonic distance sensor is installed near the roller screen inlet to better achieve the monitoring effect; the inclination sensor is installed on the inner screen frame to detect the pitch angle of the inner screen frame relative to the outer screen frame; there are four displacement sensors, which are installed next to the hydraulic cylinders of the swing control device and the pitch control device to clarify whether the control is completed; the rotary encoder is installed at the end of the eccentric screen roller assembly away from the drive assembly to monitor the real-time rotation speed of each eccentric screen roller assembly.

[0053] The communication unit includes industrial Ethernet, chip RS485, and chip RS232. Using industrial Ethernet and standard communication protocol as the communication interface, with the control system as the intermediary, the cross-type fine-grain roller screen screening parameter transmission and data interaction between each unit are completed through the industrial standard communication protocol.

[0054] In one embodiment, when the cross-type fine-grain roller screen screening parameter optimization system based on IGWO-BP-PID is working, the workflow is as follows: Figure 2 As shown, firstly, the roller screen action is obtained, and the linear array laser sensor collects the material load distribution in the opposite direction of the screen width and the ultrasonic distance sensor collects the material thickness distribution in the screen length direction, that is, the material flow information, to determine whether the roller screen is in the optimal screening condition; the linear array laser sensor and the ultrasonic distance sensor are used to determine whether the roller screen is in an optimal screening condition. When the linear array laser sensor collects the material information on the screen surface, the sensor first scans the material pile, and uses the laser reflection time and laser propagation speed of the laser beam emitted by the linear array laser sensor to the material pile to obtain a large amount of point cloud data, followed by processing the obtained point cloud data, including removing noise points and point cloud filtering. Then, the surface model of the point cloud is fitted by the least squares method to complete the recognition of the material shape, and finally the surface polynomial coefficients are extracted to obtain the mathematical model of the surface. When the ultrasonic distance measuring sensor collects the material thickness information on the screen surface, its principle is to calculate the material thickness through the time difference between the sensor emitting a short ultrasonic pulse and the sensor receiving the ultrasonic signal reflected from the coal particles on the screen surface. The collected data is first processed, and then the data is denoised, filtered, and averaged multiple times. Finally, the host computer processes the data collected by the sensor. When the material flow on the screen surface is unevenly distributed, the pitch control device and the swing control device drive the inner and outer screen frames to adjust the angles in two directions, thereby ensuring that the roller screen is always in an optimal screening condition.

[0055] Specifically, monitoring whether the screening parameters of the cross-type fine-grain roller screen meet the optimal screening conditions includes: establishing a mathematical model of the stockpile; obtaining the screening parameters of the cross-type fine-grain roller screen from the mathematical model and judging whether the optimal screening conditions are met. Among them, establishing the mathematical model of the stockpile includes: scanning the stockpile with a linear array sensor to obtain point cloud data of the stockpile; removing noise points in the point cloud data and performing point cloud filtering; using the least squares method to fit the surface model of the point cloud data to realize the recognition of the stockpile shape; extracting the polynomial coefficients of the surface model to obtain the mathematical model of the surface model.

[0056] The filtering method is statistical filtering, including calculating the distance of each point to its neighboring points to determine whether the point is a noise point. i =(x i ,yi ,z i ), the average distance d of its neighboring points m for:

[0057]

[0058] Where k is the number of neighboring points, d(p i ,p j ) indicates p i With neighboring point p j Then, the surface model of the point cloud data is fitted by the least squares method to realize the morphological recognition of the stockpile; specifically, when the surface model is fitted by the least squares method, the point cloud data is approximated by a quadratic surface equation, and when the stockpile is scanned by a linear array sensor, the three-dimensional point cloud data (x i ,y i ,z i ), and the quadratic surface equation z=ax is obtained by fitting 2 +by 2 +cxy+dx+ey+f, where a, b, c, d, e, and f are fitting coefficients; the specific steps for determining the fitting coefficients are:

[0059] Define the error function:

[0060] Construct the sum of squared errors: Where N is the total number of points,

[0061] Construct a linear equation system: Z = Xβ, where

[0062]

[0063] The fitting parameter β is obtained by solving the following normal equation:

[0064] β=(X T X) -1 X T Z, where X T is the transpose of the matrix X, (X T X) -1 Yes X T The inverse matrix of X; extract the fitting coefficients and obtain the mathematical model of the surface model.

[0065] Adjusting the mechanical system actions in the cross-type fine-grained roller screen includes: scanning the material pile through a linear array sensor to obtain the point cloud data of the material pile for swing control; obtaining the material thickness at the inlet through an ultrasonic ranging sensor for pitch control; the material thickness is calculated by the time difference between the sensor emitting a short ultrasonic pulse and the sensor receiving the ultrasonic signal reflected from the coal particles on the screen surface. The specific calculation formula is:

[0066]

[0067] v=331.4+0.6×F

[0068] Wherein, S is the distance between the ultrasonic ranging sensor and the pile, v is the speed of sound, the time difference is K, and the real-time temperature is F.

[0069] After the data is processed, it is filtered and optimized, including data denoising and filtering and averaging of multiple groups of measurements. The final output data is the point cloud data of the pile. In addition, denoising includes: when processing the point cloud data, since a large amount of dust and other interference scanning results will be generated when the cross screen is working, the noise points must be removed to obtain the initial screen surface points. Then the initial screen surface triangulation network is constructed, and the distance between the screen surface material particles and the screen surface triangulation network is calculated to determine whether the point cloud data is a noise point or a screen surface material point. Noise points are removed, and screen surface material points are added to the screen surface point set.

[0070] Specifically, after the data is preprocessed, in order to ensure the accuracy of the screen surface point cloud data, the point cloud data needs to be denoised and filtered. The first step is to remove the initial noise points. When denoising the point cloud data, the point cloud data collected initially must first be screened. This step can remove a large number of noise points that deviate significantly from the screen surface by counting the neighborhood density and distance. For each point, the neighborhood density within a certain range around it is calculated, and points with too low density are noise points. For each point p i =(x i ,y i ,z i ), we calculate the number of its neighborhood points M within a certain radius r. If the number is less than a certain threshold 5, the point is determined to be a noise point. After preliminary denoising, we construct a triangulated network for the remaining point cloud data to form a rough reference frame for the screen surface. The construction of the screen surface triangulation network can be achieved through the Delaunay triangulation algorithm. Then, by calculating the distance between the screen surface material particles and the screen surface triangulation network, it is determined whether the point cloud data is a noise point or a screen surface material point. Noise points are removed, and screen surface material points are added to the screen surface point set. Finally, after the point cloud data denoising and screening are completed, the obtained screen surface material points are further optimized. The point cloud data that has undergone multiple denoising, screening and optimization is the screen surface morphology information of the material pile, which is used for subsequent screen surface adjustment and control.

[0071] Specifically, the optimized screening parameters obtained by optimizing the preset IGWO-BP-PID algorithm include: establishing a corresponding system model according to the motion analysis of the cross-type fine-grain roller screen; designing a PID controller according to the system model; initializing the BP neural network, predicting the control output based on the historical data of the mechanical system of the cross-type fine-grain roller screen; generating an initial PID parameter population using the IGWO algorithm, each group of parameters representing the K of the PID controller p , K i , K d , where K p , K i , K d They are the proportional, integral and differential coefficients of the PID controller respectively; the fitness value of each group of PID parameters in the initial PID parameter population is calculated by using the initialized BP neural network; the population position is updated based on the fitness value, and the optimal PID parameter combination is output as the optimized screening parameter to control the mechanical system. At the same time, in order to avoid the initialization population distribution being too concentrated or falling into the local optimum, chaotic mapping is used instead of traditional random initialization to generate the initial PID parameter population.

[0072] Specifically, the motion analysis of the cross-type fine-grain roller screen includes establishing the following motion model:

[0073]

[0074] Among them, θ1 is the initial pitch angle; θ2 is the angle of pitch adjustment, that is, the angle changed by the pitch hydraulic cylinder after the pitch control device is operated (the pitch hydraulic cylinder is extended or retracted); θ1+θ2 is the angle after pitch control, β1 is the initial swing angle; β2 is the angle of swing adjustment, that is, the angle changed by the swing hydraulic cylinder after the swing control device is operated (the swing hydraulic cylinder is extended or retracted); β1+β2 is the angle after swing control, x is the extension length of the hydraulic cylinder in the swing control device , y is the extension length of the hydraulic cylinder in the pitch control device, A is the distance between the rotation center of the inner and outer frame connection device and the same plane as the hinge point of the pitch hydraulic cylinder and the top plate, B is the distance between the rotation center of the inner and outer frame connection device and the pitch hydraulic cylinder connection seat and the inner frame hinge point, C is the initial length of the pitch hydraulic cylinder, a is the distance from the front support frame hinge point to the support swing hydraulic cylinder and the ground contact support frame hinge point, b is the distance from the front support frame hinge point to the swing hydraulic cylinder and the swing hydraulic cylinder connection seat hinge point, c is the initial height of the swing hydraulic cylinder. In this embodiment, A, B, C, a, b, c can be respectively from Figure 3 and Figure 4 Measured on the crossed fine roller screen shown.

[0075] Specifically, in a mechanical system, the PID parameter combination is used to adjust the extension of the hydraulic cylinder to achieve control of the mechanical system;

[0076] Adjusting the extension x(t) of the hydraulic cylinder involves making adjustments using the following formula:

[0077]

[0078] Where e(t) = x setpoint -x(t),K p , K i , K d are the proportional, integral, and differential coefficients of the PID controller, u(t) represents the PID output, x(t) is the actual extension of the hydraulic cylinder, and x setpoint represents the target elongation of the hydraulic cylinder, e(t) represents the difference between the actual elongation and the target elongation, t is the time variable representing the elongation process, e(t) = x setpoint -x(t),K p , K i , K d are the proportional, integral, and differential coefficients of the PID controller, respectively, and e(τ) represents the value of the cumulative elongation error at different time points. By integrating the error over the entire time period (i.e. calculating the cumulative error), the PID controller can effectively adjust the system response to make the actual output closer to the set value.

[0079] The fitness value is defined based on the performance indicators of the hydraulic cylinder:

[0080] Fitness=w1×ITAE+w2×MSE+w3×Overshoot

[0081]

[0082] Among them, the integral time absolute error ITAE is used to measure the error and response time of the hydraulic cylinder, the mean square error MSE is the steady-state error, and Overshoot is the overshoot, which is used to indicate the maximum deviation of the elongation of the hydraulic cylinder before reaching the set value; w1, w2, and w3 are weight coefficients, w1 corresponds to ITAE, which measures the response performance of the system in time and reflects the tracking performance of the system to the set value; this weight is usually used to emphasize the importance of reducing the error in time during the control process; w2 corresponds to MSE, which measures the average error between the system output and the expected output, reflecting the control accuracy. This weight is used to focus on the steady-state accuracy of the system and ensure that the system output is as close to the expected value as possible; w3 corresponds to Overshoot, and the overshoot refers to the degree to which the system output exceeds the value when it reaches the steady-state value, reflecting the dynamic characteristics of the system. This weight is used to control the response speed and stability of the system to ensure that the system does not fluctuate excessively; T corresponds to the total time period or the total number of sampling points for calculating the error. After the mechanical system is controlled by adjusting the elongation of the hydraulic cylinder through the optimized PID parameter combination, the inclination of the screen surface can be adjusted, which is conducive to the smooth progress of the screening operation. And in this system, the adjustment process can be fully automatic, further improving efficiency.

[0083] The embodiments of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited thereto. Various changes that can be made within the knowledge scope of technicians in the relevant technical field without departing from the spirit of the present invention are all within the protection scope of the claims of the present invention.

[0084] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A cross-screen screening parameter optimization system based on IGWO-BP-PID, characterized in that: It includes a cross-type fine-grain roller screen, a control system, a communication unit, a monitoring system, and a host computer; The monitoring system is connected to the host computer via the communication unit, and the monitoring system is used to monitor whether the screening parameters of the cross-type fine-grain roller screen meet the optimal screening conditions; When the optimal screening condition is not met, the monitoring system transmits the screening parameter to the host computer through the communication unit; The host computer is also connected to the control system. The host computer reads the screening parameters and optimizes them through a preset IGWO-BP-PID algorithm to obtain optimized screening parameters, and transmits the optimized screening parameters to the control system; The control system is also connected to the cross-type fine-grain roller screen, and the control system is used to adjust the mechanical system action in the cross-type fine-grain roller screen according to the optimized screening parameters so that it meets the optimal screening conditions.

2. A cross-screen screening parameter optimization method based on IGWO-BP-PID, applied to the host computer, characterized in that: include: Receiving screening parameters sent by the monitoring system that do not meet the optimal screening conditions; The screening parameters are optimized by using a preset IGWO-BP-PID algorithm to obtain optimized screening parameters; The optimized screening parameters are transmitted to a control system.

3. The cross-screen screening parameter optimization method based on IGWO-BP-PID according to claim 2 is characterized in that: The screening parameters are optimized by the preset IGWO-BP-PID algorithm to obtain the optimized screening parameters including: Establishing a corresponding motion model based on the motion analysis of the cross-type fine-grain roller screen; Designing a PID controller based on the system model; Initializing a BP neural network to predict control output based on historical data of the mechanical system of the cross-type fine-grain roller screen; The IGWO algorithm is used to generate the initial PID parameter population, each group of parameters represents the K of the PID controller. p , K i , K d , where K p , K i , K d They are the proportional, integral and differential coefficients of the PID controller respectively; The initialized BP neural network is used to calculate the fitness value of each group of PID parameters in the initial PID parameter population; the population position is updated based on the fitness value, and the optimal PID parameter combination is output as the optimized screening parameter to control the mechanical system.

4. The cross-screen screening parameter optimization method based on IGWO-BP-PID according to claim 3 is characterized in that: The method of generating an initial PID parameter population using the IGWO algorithm includes: A chaos map is used to generate the initial PID parameter population.

5. The cross-screen screening parameter optimization method based on IGWO-BP-PID according to claim 3, characterized in that: In the mechanical system, the PID parameter combination adjusts the elongation of the hydraulic cylinder through the following formula to achieve control of the mechanical system; Adjusting the extension x(t) of the hydraulic cylinder involves making adjustments using the following formula: Where u(t) represents the PID output, x(t) is the actual extension of the hydraulic cylinder, and x setpoint represents the target elongation of the hydraulic cylinder, e(t) represents the difference between the actual elongation and the target elongation, t is the time variable representing the elongation process, e(t) = x setpoint -x(t),K p , K i , K d are the proportional, integral, and differential coefficients of the PID controller, respectively, and e(t) represents the value of the cumulative elongation error at different time points; The fitness value is defined based on the performance index of the hydraulic cylinder: Fitness=w1×ITAE+w2×MSE+w3×Overshoot Among them, the integral of time-weighted absolute error (ITAE) is a measure of the error and response time of the hydraulic cylinder, the mean square error (MSE) is the square average of the error; Overshoot is the overshoot, which is used to indicate the maximum deviation of the elongation of the hydraulic cylinder before reaching the set value; w1, w2, and w3 are weight coefficients, w1 corresponds to ITAE, w2 corresponds to MSE, and w3 corresponds to Overshoot; T corresponds to the total number of time periods or the total number of sampling points for calculating the error.

6. The cross-screen screening parameter optimization method based on IGWO-BP-PID according to claim 3, characterized in that: The motion analysis of the cross-type fine-grain roller screen includes establishing the following motion model: Among them, θ1 is the initial pitch angle, θ2 is the pitch adjustment angle, θ1+θ2 is the angle after pitch control, β1 is the initial swing angle, β2 is the swing adjustment angle, β1+β2 is the angle after swing control, x is the extension length of the hydraulic cylinder in the swing control device, y is the extension length of the hydraulic cylinder in the pitch control device, A is the distance between the rotation center of the inner and outer frame connecting device and the same plane as the pitch hydraulic cylinder and the top plate hinge point, B is the distance between the rotation center of the inner and outer frame connecting device and the pitch hydraulic cylinder connecting seat and the inner frame hinge point, C is the initial length of the pitch hydraulic cylinder, a is the distance from the front support frame hinge point to the support swing hydraulic cylinder and the ground contact support frame hinge point, b is the distance from the front support frame hinge point to the swing hydraulic cylinder and the swing hydraulic cylinder connecting seat hinge point, c is the initial height of the swing hydraulic cylinder.

7. The cross-screen screening parameter optimization system based on IGWO-BP-PID according to claim 1 is characterized in that: Monitoring whether the screening parameters of the cross-type fine-grain roller screen meet the optimal screening conditions includes: Build a mathematical model of the stockpile; The screening parameters of the cross-type fine-grain roller screen are obtained from the mathematical model to determine whether the optimal screening conditions are met.

8. The cross-screen screening parameter optimization system based on IGWO-BP-PID according to claim 7, characterized in that: The mathematical model of the stockpile is established, comprising: Scanning a pile of materials by means of a linear array sensor to obtain point cloud data of the pile of materials; Removing noise points from the point cloud data and performing point cloud filtering processing; The filtering processing method is statistical filtering, including determining whether a point is a noise point by calculating the distance of each point to its neighboring points. i =(x i ,y i ,z i ), the average distance d of its neighboring points m for: Where k is the number of neighboring points, d(p i ,p j ) indicates p i With neighboring point p j distance; Then, the surface model of the point cloud data is fitted using the least square method to realize the morphological recognition of the stockpile; Specifically, when the least squares method is used to fit the surface model, a quadratic surface equation is used to approximate the point cloud data, and when the pile is scanned by a linear array sensor, three-dimensional point cloud data (x i ,y i ,z i ), and the quadratic surface equation z=ax is obtained by fitting 2 +by 2 +cxy+dx+ey+f, where a, b, c, d, e, and f are fitting coefficients; the specific steps for determining the fitting coefficients are: Define the error function: Construct the sum of squared errors: Where N is the total number of points, Construct a linear equation system: Z = Xβ, where The fitting parameter β is obtained by solving the following normal equation: β=(X T X) -1 X T Z, where X T is the transpose of the matrix X, (X T X) -1 Yes X T The inverse matrix of X; The fitting coefficients are extracted to obtain a mathematical model of the surface model.

9. The cross-screening parameter optimization system based on IGWO-BP-PID according to claim 8, characterized in that: Adjustment of the mechanical system actions in the cross-type fine-grain roller screen includes: Scanning the pile of materials through a linear array sensor to obtain point cloud data of the pile of materials for swing control; The material thickness at the inlet is obtained by the ultrasonic distance sensor for pitch control; the material thickness is calculated by the time difference between the ultrasonic distance sensor emitting a short ultrasonic pulse and the ultrasonic distance sensor receiving the ultrasonic signal reflected from the screen material pile. The specific calculation formula is: v=331.4+0.6×F Wherein, S is the distance between the ultrasonic ranging sensor and the pile, v is the speed of sound, the time difference is K, and the real-time temperature is F.

Images