Screening efficiency optimization method applied to vibrating screen

Abstract

The invention relates to a screening efficiency optimization method applied to a vibrating screen, belongs to the technical field of screening efficiency optimization, and solves the problems that a method for determining screening parameters of a vibrating screen in the prior art is poor in adaptability, low in accuracy and incapable of adapting to complex conditions. Obtaining a to-be-tested input parameter corresponding to the current to-be-tested vibrating screen, and inputting the to-be-tested input parameter into the target screening efficiency prediction model to obtain the predicted screening efficiency of the to-be-tested vibrating screen; and if the predicted screening efficiency meets an expected target, taking the to-be-tested input parameter as a control parameter of the to-be-tested vibrating screen, and if the predicted screening efficiency does not meet the expected target, optimizing the to-be-tested input parameter by adopting a genetic algorithm, and taking the optimized parameter as the control parameter of the to-be-tested vibrating screen. The screening efficiency of the to-be-tested vibrating screen is improved; in the optimization process, input parameters represented by each individual are input into the target screening efficiency prediction model to obtain the screening efficiency of each individual, and a fitness function is obtained based on the screening efficiency, energy consumption and production cost of each individual. The method for reasonably, rapidly and efficiently determining the screening parameters of the vibrating screen is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of screening efficiency optimization, and in particular to a screening efficiency optimization method applied to a vibrating screen. Background Art

[0002] Vibrating screen is a screening equipment widely used in industrial production. It is mainly used for grading, screening, filtering and other operations of materials. It uses the vibration force generated by the vibration motor or exciter to screen the materials on the screen surface, thereby achieving particle grading and screening. The screening efficiency of the vibrating screen is an important indicator to measure its working effect. It reflects the ability of the vibrating screen to screen materials in actual operation. In actual production, in order to obtain higher screening efficiency, it is necessary to adjust the parameters such as feeding speed, feeding amount, feeding method, screen surface inclination, feed moisture, and working mode of the vibration motor to meet the needs of different materials.

[0003] In the prior art, the adjustment of relevant parameters of the vibrating screen often relies on experienced technicians, and the manual setting method has the following disadvantages: (1) Different operators may have different setting habits and standards, resulting in difficulty in ensuring the consistency and accuracy of the screening results; (2) It is difficult for operators to adjust parameters in a timely manner according to actual conditions, which may cause the equipment to operate under undesirable parameters and affect the screening efficiency and quality; (3) In a complex production environment, such as processing a variety of materials of different properties or under harsh working conditions, it is difficult for manually set parameters to quickly adapt to changes, which may result in poor screening results.

[0004] Therefore, the prior art lacks a method for reasonably, quickly and efficiently determining the screening parameters of the vibrating screen. Summary of the invention

[0005] In view of the above analysis, an embodiment of the present invention aims to provide a screening efficiency optimization method applied to a vibrating screen, so as to solve the problems that the existing method for determining the screening parameters of the vibrating screen has poor adaptability, low accuracy and cannot adapt to complex situations.

[0006] An embodiment of the present invention provides a screening efficiency optimization method applied to a vibrating screen, comprising:

[0007] Acquire historical screening data of the vibrating screen, and perform preprocessing operations on the historical screening data to form a training sample set;

[0008] The screening efficiency prediction model CNN-GRU is trained based on the training sample set to obtain a target screening efficiency prediction model that meets preset conditions;

[0009] Obtaining the input parameters to be tested corresponding to the current vibrating screen to be tested, and inputting the parameters into the target screening efficiency prediction model to obtain the predicted screening efficiency of the vibrating screen to be tested;

[0010] If the predicted screening efficiency meets the expected target, the input parameter to be tested is used as the control parameter of the vibrating screen to be tested; if it does not meet the expected target, a genetic algorithm is used to optimize the input parameter to be tested, and the optimized parameter is used as the control parameter of the vibrating screen to be tested, thereby improving the screening efficiency of the vibrating screen to be tested;

[0011] During the optimization process, the input parameters represented by each individual are input into the target screening efficiency prediction model to obtain the screening efficiency of each individual, and the fitness function is obtained based on the screening efficiency, energy consumption and production cost of each individual.

[0012] Based on further improvement of the above method, the historical screening data includes the vibration signal, vibration frequency, exciting force, screen hole size, screen surface inclination, as well as material particle size, material moisture, feed speed, and screening efficiency of the vibrating screen.

[0013] Based on the further improvement of the above method, the preprocessing operation includes: processing the vibration signal based on the empirical mode decomposition EMD algorithm and the Pearson correlation coefficient PCC algorithm to obtain the inherent mode function IMF that meets the requirements, reconstructing all the inherent mode functions IMF that meet the requirements to obtain the first characteristic component; obtaining multiple window data in the first characteristic component based on the sliding window method, and splicing the multiple window data to obtain the second characteristic component; normalizing the material particle size and the material moisture to obtain the first material particle size and the first material moisture; using the second characteristic component, the first material particle size, the first material moisture, the vibration frequency, the exciting force, the screen hole size, the screen surface inclination, the feed speed, and the screening efficiency as sample data to form a training sample set.

[0014] Based on the further improvement of the above method, the screening efficiency prediction model CNN-GRU is trained based on the training sample set to obtain a target screening efficiency prediction model that meets preset conditions, including: taking the screening efficiency in the training sample set as the true value and the remaining parameters as input data, training the screening efficiency prediction model CNN-GRU, and when the change value of the loss function value of the screening efficiency prediction model CNN-GRU in multiple consecutive iterations is less than a preset threshold or reaches the upper limit of the number of iterations, the target screening efficiency prediction model is obtained.

[0015] Based on the further improvement of the above method, the loss function of the screening efficiency prediction model CNN-GRU is for:

[0016]

[0017] in, is the root mean square error, is the mean absolute error, Smoothing loss, α, β, δ are constants obtained based on trial and error.

[0018] Based on the further improvement of the above method, the smoothing loss is obtained as follows:

[0019]

[0020] Among them, y i is the screening efficiency output of the i-th iteration, y i-1 is the screening efficiency output of the i-1th iteration, y i-2 is the screening efficiency output of the i-2th iteration, μ is the smoothing control coefficient, 0<μ<1, i is the number of iterations, i=3,4,5,…,M.

[0021] Based on the further improvement of the above method, the vibration signal is collected based on a sensor installed at a specified position, and the specified position includes a screen box, a screen mesh, and a bracket of the vibrating screen.

[0022] Based on the further improvement of the above method, the vibration signal is processed based on the empirical mode decomposition EMD algorithm and the Pearson correlation coefficient PCC algorithm to obtain the intrinsic mode function IMF that meets the requirements, and all the intrinsic mode functions IMF that meet the requirements are reconstructed to obtain the first characteristic component, including:

[0023] S21: performing an empirical mode decomposition (EMD) algorithm on the acquired vibration signal to generate an IMF component;

[0024] S22: calculating the Pearson correlation coefficient PCC between each IMF component and its corresponding vibration signal, and screening the IMF components whose Pearson correlation coefficient PCC is greater than a first preset threshold to form an IMF component set;

[0025] S23: Sort the components in the IMF component set in descending order according to the Pearson correlation coefficient PCC, and assign larger weights to the IMF components with higher rankings, and smaller weights to the components with lower rankings. The sum of the weights corresponding to the components is 1, and the weighted sum of the components and their corresponding weights is taken as the first characteristic component.

[0026] Based on the further improvement of the above method, the genetic algorithm is used to optimize the input parameters to be tested, and the optimized parameters are used as the control parameters of the vibrating screen to be tested, including:

[0027] A1: based on the predicted screening efficiency, sample data of the same type as the current vibrating screen are selected from the training sample set to form a first similar data set, the difference between the screening efficiency in any sample data in the first similar data set and the predicted screening efficiency is a positive number, and the input parameter to be tested and the input parameter data of each sample in the first similar data set are used as individuals of the initial population;

[0028] A2: If the number of individuals in the current initial population meets the preset number requirement, execute step A4; otherwise, execute A3;

[0029] A3: Based on the random generation algorithm, the values ​​of each input parameter are selected within the valid range to generate individuals, so that the number of individuals meets the preset number requirements;

[0030] A4: Use the target screening efficiency prediction model to predict the screening efficiency of each individual in the initial population. After obtaining the screening efficiency, calculate the fitness function value of each individual.

[0031] A5: Use the tournament selection algorithm to select individuals with large fitness function values ​​from the current population to enter the next generation;

[0032] A6: Perform crossover and mutation operations to generate a new generation of population;

[0033] A7: Determine whether the current iteration meets the preset termination condition. If so, the process ends; otherwise, the process returns to A5.

[0034] A8: Select the individual with the largest fitness function value as the optimized parameter, and set it as the control parameter of the vibrating screen to be tested.

[0035] Based on the further improvement of the above method, the fitness evaluation function F(x) is:

[0036] F(x)=w1y1-w2y2-w3y3,

[0037] Among them, y1, y2, and y3 represent screening efficiency, energy consumption, and production cost respectively, and w1, w2, and w3 are weight coefficients.

[0038] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0039] 1. The present invention provides a screening efficiency optimization method for a vibrating screen. A target screening efficiency prediction model and a corresponding input parameter combination are obtained based on historical screening sample data. After obtaining the input parameters to be tested of the current vibrating screen to be tested, the screening efficiencies of different types of vibrating screens can be obtained. Therefore, the present invention provides a standardized method for predicting the screening efficiency of a vibrating screen, which can be adapted to different types of vibrating screens. For the input parameters to be tested in different scenarios, the screening efficiency of the vibrating screen can be obtained accurately and quickly.

[0040] 2. The present invention provides a screening efficiency optimization method for a vibrating screen. After the screening efficiency is calculated based on the input parameters to be tested of the current vibrating screen to be tested, if the screening efficiency requirement is not met, the genetic algorithm can be used to obtain the input parameters that meet the screening efficiency. Therefore, the present invention reduces the difficulty of adjusting the input parameters, and does not require relevant professionals to make adjustments based on professional knowledge. Only relying on the provided genetic algorithm, the optimized input parameters that match the current vibrating screen to be tested and the input parameters can be quickly obtained.

[0041] 3. The present invention provides a screening efficiency optimization method for a vibrating screen. When processing the vibration signal, the sliding window method is used to increase the dimension of the vibration signal, thereby increasing the feature expression capability of the training sample set. The loss function in the genetic algorithm takes into account the effects of screening efficiency, energy consumption, and production cost, further improving the effectiveness of the results.

[0042] In the present invention, the above-mentioned technical solutions can also be combined with each other to achieve more preferred combination solutions. Other features and advantages of the present invention will be described in the subsequent description, and some advantages can become obvious from the description, or can be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained through the contents particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings are only used for the purpose of illustrating specific embodiments and are not to be considered as limiting the present invention. In the entire drawings, the same reference symbols represent the same components;

[0044] Figure 1 This is a flowchart of a screening efficiency optimization method applied to a vibrating screen in an embodiment of the present invention;

[0045] Figure 2 This is an example diagram of the process of obtaining the first characteristic component in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0047] A specific embodiment of the present invention discloses a method for optimizing the screening efficiency of a vibrating screen, such as Figure 1 As shown, including:

[0048] S1: Acquire historical screening data of the vibrating screen, and perform preprocessing operations on the historical screening data to form a training sample set.

[0049] Historical screening data may include the vibration signal, vibration frequency, exciting force, screen hole size, screen surface inclination, as well as material particle size, material moisture, feed rate, screening efficiency, feed amount, opening rate, bulk density, execution time, etc.

[0050] The vibration signal is collected based on sensors installed at designated locations, including the screen box, screen mesh, and bracket of the vibrating screen. The present invention does not limit the number and location of collected vibration signals, and only requires meeting actual needs.

[0051] The vibration frequency and exciting force are directly determined by the type of vibrating screen. Parameters such as screen hole size, screen surface inclination, material particle size, material moisture, feed speed, feed amount, opening rate, bulk density, etc. can be determined according to the actual scenario.

[0052] It is understandable that the screening efficiency of the vibrating screen is closely related to the vibration signal. Therefore, the more vibration signals collected, the more accurate the prediction result of the final screening efficiency. Preferably, the present invention collects 4 vibration signals, including 1 screen box, 1 screen mesh, and 2 brackets, which are respectively located at the top and the middle of the bracket.

[0053] It is understandable that there are many parameters that affect the screening efficiency of the vibrating screen, some parameters are difficult to obtain, and some parameters have little effect on the screening efficiency. Therefore, the present invention statistically organizes the parameters that affect the screening efficiency, and selects one of the parameter screening methods such as principal component analysis PCA, factor analysis, feature selection algorithm or decision tree, and finally uses the vibration signal, vibration frequency, exciting force, screen hole size, screen surface inclination, as well as material particle size, material moisture, feed speed, and screening efficiency of the vibrating screen as historical screening data.

[0054] The preprocessing operation includes data cleaning, data conversion, data dimension reduction, data reconstruction, etc. of the historical screening data. Through preprocessing, the data quality can be improved and the accuracy of the screening efficiency prediction can be improved. It should be noted that the data preprocessing operation can be determined according to business needs, and the present invention is not limited here, and it is subject to being able to achieve business needs.

[0055] The preprocessing operation also includes processing the vibration signal based on the empirical mode decomposition EMD algorithm and the Pearson correlation coefficient PCC algorithm to obtain the required intrinsic mode function IMF, reconstructing all the required intrinsic mode functions IMF to obtain the first characteristic component; obtaining multiple window data in the first characteristic component based on the sliding window method, and splicing the multiple window data to obtain the second characteristic component. The material particle size and the material humidity are normalized to obtain the first material particle size and the first material humidity. The second characteristic component, the first material particle size, the first material humidity, the vibration frequency, the exciting force, the screen hole size, the screen surface inclination, the feed speed, and the screening efficiency are used as sample data to form a training sample set.

[0056] The method of processing the vibration signal based on the empirical mode decomposition (EMD) algorithm and the Pearson correlation coefficient (PCC) algorithm to obtain an intrinsic mode function (IMF) that meets the requirements, and reconstructing all intrinsic mode functions (IMF) that meet the requirements to obtain the first characteristic component includes: executing the empirical mode decomposition (EMD) algorithm on the acquired vibration signal to generate IMF components. Calculating the Pearson correlation coefficient (PCC) between each IMF component and its corresponding vibration signal, and screening the components whose Pearson correlation coefficient (PCC) in the IMF component is greater than the first preset threshold to form an IMF component set. Sort the components in the IMF component set in descending order according to the Pearson correlation coefficient (PCC), and assign a larger weight to the top-ranked IMF component, and a smaller weight to the bottom-ranked component, the sum of the weights corresponding to each component is 1, and the weighted sum of each component and its corresponding weight is used as the first characteristic component, as shown in 2.

[0057] For example, the first preset threshold value may be 0.7, and the first characteristic component IMf total for:

[0058] IMF total =W1*IMf1+W2*IMf2+W3*IMf3+.....+W N *IMf N ,

[0059] in,

[0060] W1+W2+W3+...+W N =1,

[0061] IMf1>IMf2>IMf3>...>IMf N ,

[0062] W1>W2>W3>...>W N ,

[0063] Among them, W N corresponds to IMfN The weight of IMf N is the component in the IMF component set, and N is the number of components in the IMF component set.

[0064] The sliding window method is used to increase the dimension of the vibration signal, which increases the feature expression ability of the training sample set.

[0065] S2: Training the screening efficiency prediction model CNN-GRU based on the training sample set to obtain a target screening efficiency prediction model that meets preset conditions.

[0066] Specifically, the screening efficiency in the training sample set is taken as the true value and the other parameters are used as input data to train the screening efficiency prediction model CNN-GRU. When the change value of the loss function value of the screening efficiency prediction model CNN-GRU in multiple consecutive iterations is less than the preset threshold or reaches the upper limit of the number of iterations, the target screening efficiency prediction model is obtained.

[0067] Specifically, after obtaining the training sample set, it is divided into a training set, a test set, and a validation set according to a certain ratio, and then the screening efficiency prediction model CNN-GRU is trained.

[0068] Preferably, the set is divided into training set, test set and validation set in a ratio of 8:1:1; the preset threshold is 0.001, and the upper limit of the number of iterations is 1000.

[0069] Among them, CNN is composed of an input layer, a convolution layer, a pooling layer, etc. The present invention does not limit the specific types of CNN and GRU models. The number of layers, activation functions and other parameters of the model in its CNN-GRU are based on being able to meet the needs. It only needs to ensure that the input layer of the CNN neural network corresponds to the input parameters in the training sample set. The last pooling layer of CNN is connected to the GRU model, and GRU is used to input the final predicted screening efficiency.

[0070] The loss function of the screening efficiency prediction model CNN-GRU for:

[0071]

[0072] in, is the root mean square error, is the mean absolute error, Smoothing loss, α, β, δ are constants obtained based on trial and error. The root mean square error is:

[0073]

[0074] The mean absolute error is:

[0075]

[0076] The smoothing loss is obtained as follows:

[0077]

[0078] Among them, y i is the screening efficiency output at the i-th iteration, is the screening efficiency recorded in the sample data set, y i-1 is the screening efficiency output of the i-1th iteration, y i-2 is the screening efficiency outputted at the i-2th iteration, μ is the smoothing control coefficient, 0<μ<1, i is the number of iterations, i=3, 4, 5, ..., M, and n is the number of samples.

[0079] Preferably, α, β, and δ may be 0.02, 0.02, and 0.1.

[0080] During the model training process, the present invention considers the impact of each iteration, and each training is adjusted according to the results of the previous and previous two iterations in the hope of reducing losses. This continuous optimization process forms a cumulative effect between the output results of each iteration. If a certain iteration finds that the model performs poorly in the classification of certain categories, more attention will be paid to these categories in the next iteration to improve the overall performance.

[0081] S3: obtaining the input parameters to be tested corresponding to the current vibrating screen to be tested, and inputting the parameters into the target screening efficiency prediction model to obtain the predicted screening efficiency of the vibrating screen to be tested;

[0082] If the predicted screening efficiency meets the expected target, the input parameter to be tested is used as the control parameter of the vibrating screen to be tested; if it does not meet the expected target, a genetic algorithm is used to optimize the input parameter to be tested, and the optimized parameter is used as the control parameter of the vibrating screen to be tested, thereby improving the screening efficiency of the vibrating screen to be tested;

[0083] During the optimization process, the input parameters represented by each individual are input into the target screening efficiency prediction model to obtain the screening efficiency of each individual, and the fitness function is obtained based on the screening efficiency, energy consumption and production cost of each individual.

[0084] The method of optimizing the input parameters to be tested by using a genetic algorithm and using the optimized parameters as control parameters of the vibrating screen to be tested includes:

[0085] A1: Based on the predicted screening efficiency, sample data of the same type as the current vibrating screen are selected from the training sample set to form a first similar data set, and the difference between the screening efficiency in any sample data in the first similar data set and the predicted screening efficiency is a positive number, and the input parameters to be tested and the input parameter data of each sample in the first similar data set are used as individuals of the initial population.

[0086] A2: If the number of individuals in the current initial population meets the preset number requirement, execute step A4; otherwise, execute A3.

[0087] A3: Based on the random generation algorithm, the values ​​of each input parameter are selected within the valid range to generate individuals so that the number of individuals meets the preset number requirements.

[0088] Preferably, the preset number is 50.

[0089] A4: The target screening efficiency prediction model is used to predict the screening efficiency of each individual in the initial population. After the screening efficiency is obtained, the fitness function value of each individual is calculated; the fitness evaluation function F(x) is:

[0090] F(x)=w1y1-w2y2-w3y3,

[0091] Among them, y1, y2, and y3 represent screening efficiency, energy consumption, and production cost respectively, and w1, w2, and w3 are weight coefficients.

[0092] Specifically, the screening efficiency corresponding to each individual can be obtained by inputting the input parameters of each individual into the target screening efficiency prediction model.

[0093] In addition, the historical screening data also includes the rated current, rated voltage and execution time of this type of vibrating screen, as well as cost records (material loss costs, equipment maintenance costs and labor costs, etc.).

[0094] Taking the current individual as the first vector to be tested, a sample data closest to the current first vector to be tested is selected from the historical screening data, the energy consumption is calculated using the rated current, rated voltage and execution time recorded in the sample data, and the production cost is obtained by accumulating the expenses in the cost record in the sample data.

[0095] The method of selecting a piece of sample data closest to the current first vector to be tested from the historical screening data may adopt the nearest neighbor method, cosine similarity, KD-Tree and BST, ElasticSearch algorithm, and the present invention is not limited thereto, and the main focus is on obtaining a piece of sample data that is most similar to the current one.

[0096] The weight coefficients can be obtained according to actual needs or by trial and error. Preferably, w1, w2, and w3 are 0.5, 0.3, and 0.15, respectively.

[0097] A5: Use the tournament selection algorithm to select individuals with large fitness function values ​​from the current population to enter the next generation.

[0098] A6: Perform crossover and mutation operations to generate a new generation of population.

[0099] A7: Determine whether the current iteration meets the preset termination condition. If so, the process ends; otherwise, the process returns to A5.

[0100] A8: Select the individual with the largest fitness function value as the optimized parameter, and set it as the control parameter of the vibrating screen to be tested.

[0101] It can be understood that the training and implementation process of the genetic algorithm are common knowledge in the art, and the present application does not make any special limitation thereto, but only applies the genetic algorithm to optimize the input parameters.

[0102] Compared with the prior art, the present embodiment provides a screening efficiency optimization method for a vibrating screen, which can be adapted to different types of vibrating screens and can accurately and quickly obtain the vibrating screen screening efficiency for input parameters to be tested in different scenarios; it reduces the difficulty of adjusting the input parameters, and does not require relevant professionals to make adjustments based on professional knowledge. It only relies on the provided genetic algorithm to quickly obtain the optimized input parameters that match the current vibrating screen to be tested and the input parameters; when processing the vibration signal, the sliding window method is used to increase the dimension of the vibration signal, thereby increasing the feature expression capability of the training sample set.

[0103] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0104] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A screening efficiency optimization method for a vibrating screen, characterized in that: include: Acquire historical screening data of the vibrating screen, and perform preprocessing operations on the historical screening data to form a training sample set; The screening efficiency prediction model CNN-GRU is trained based on the training sample set to obtain a target screening efficiency prediction model that meets preset conditions; Obtaining the input parameters to be tested corresponding to the current vibrating screen to be tested, and inputting the parameters into the target screening efficiency prediction model to obtain the predicted screening efficiency of the vibrating screen to be tested; If the predicted screening efficiency meets the expected target, the input parameter to be tested is used as the control parameter of the vibrating screen to be tested; if it does not meet the expected target, a genetic algorithm is used to optimize the input parameter to be tested, and the optimized parameter is used as the control parameter of the vibrating screen to be tested, thereby improving the screening efficiency of the vibrating screen to be tested; During the optimization process, the input parameters represented by each individual are input into the target screening efficiency prediction model to obtain the screening efficiency of each individual, and the fitness function is obtained based on the screening efficiency, energy consumption and production cost of each individual.

2. The screening efficiency optimization method for a vibrating screen according to claim 1, characterized in that: The historical screening data include the vibration signal, vibration frequency, exciting force, screen hole size, screen surface inclination, as well as material particle size, material moisture, feed speed, and screening efficiency of the vibrating screen.

3. The screening efficiency optimization method for a vibrating screen according to claim 2, characterized in that: The pre-processing operation includes: The vibration signal is processed based on the empirical mode decomposition (EMD) algorithm and the Pearson correlation coefficient (PCC) algorithm to obtain an intrinsic mode function (IMF) that meets the requirements, and all intrinsic mode functions (IMF) that meet the requirements are reconstructed to obtain a first characteristic component; multiple window data in the first characteristic component are obtained based on a sliding window method, and the multiple window data are spliced ​​to obtain a second characteristic component; Normalizing the material particle size and the material humidity to obtain a first material particle size and a first material humidity; The second characteristic component, the particle size of the first material, the humidity of the first material, the vibration frequency, the exciting force, the screen hole size, the screen surface inclination, the feed speed, and the screening efficiency are used as sample data to form a training sample set.

4. The screening efficiency optimization method for a vibrating screen according to claim 3, characterized in that: The screening efficiency prediction model CNN-GRU is trained based on the training sample set to obtain a target screening efficiency prediction model that meets preset conditions, including: The screening efficiency in the training sample set is taken as the true value and the other parameters are taken as input data to train the screening efficiency prediction model CNN-GRU. When the change value of the loss function value of the screening efficiency prediction model CNN-GRU in multiple consecutive iterations is less than the preset threshold or reaches the upper limit of the iteration number, the target screening efficiency prediction model is obtained.

5. The screening efficiency optimization method for a vibrating screen according to claim 4, characterized in that: The loss function of the screening efficiency prediction model CNN-GRU for: in, is the root mean square error, is the mean absolute error, Smoothing loss, α, β, δ are constants obtained based on trial and error.

6. A screening efficiency optimization method for a vibrating screen according to claim 5, wherein: The smoothing loss is obtained as follows: Among them, yi is the screening efficiency output of the i-th iteration, y i-1 is the screening efficiency output of the i-1th iteration, y i-2 is the screening efficiency output of the i-2th iteration, μ is the smoothing control coefficient, 0<μ<1, i is the number of iterations, i=3,4,5,…,M.

7. A screening efficiency optimization method for a vibrating screen according to any one of claims 1 to 6, characterized in that: The vibration signal is collected based on a sensor installed at a specified position, and the specified position includes a screen box, a screen mesh, and a bracket of the vibrating screen.

8. The screening efficiency optimization method for a vibrating screen according to claim 3, characterized in that: The vibration signal is processed based on the empirical mode decomposition (EMD) algorithm and the Pearson correlation coefficient (PCC) algorithm to obtain an intrinsic mode function (IMF) that meets the requirements, and all intrinsic mode functions (IMF) that meet the requirements are reconstructed to obtain a first characteristic component, including: S21: performing an empirical mode decomposition (EMD) algorithm on the acquired vibration signal to generate an IMF component; S22: calculating the Pearson correlation coefficient PCC between each IMF component and its corresponding vibration signal, and screening the IMF components whose Pearson correlation coefficient PCC is greater than a first preset threshold to form an IMF component set; S23: Sort the components in the IMF component set in descending order according to the Pearson correlation coefficient PCC, and assign larger weights to the IMF components with higher rankings, and smaller weights to the components with lower rankings. The sum of the weights corresponding to the components is 1, and the weighted sum of the components and their corresponding weights is taken as the first characteristic component.

9. The screening efficiency optimization method for a vibrating screen according to claim 8, characterized in that: The method of optimizing the input parameters to be tested by using a genetic algorithm and using the optimized parameters as control parameters of the vibrating screen to be tested includes: A1: based on the predicted screening efficiency, sample data of the same type as the current vibrating screen are selected from the training sample set to form a first similar data set, the difference between the screening efficiency in any sample data in the first similar data set and the predicted screening efficiency is a positive number, and the input parameter to be tested and the input parameter data of each sample in the first similar data set are used as individuals of the initial population; A2: If the number of individuals in the current initial population meets the preset number requirement, execute step A4; otherwise, execute A3; A3: Based on the random generation algorithm, the values ​​of each input parameter are selected within the valid range to generate individuals, so that the number of individuals meets the preset number requirements; A4: Use the target screening efficiency prediction model to predict the screening efficiency of each individual in the initial population. After obtaining the screening efficiency, calculate the fitness function value of each individual. A5: Use the tournament selection algorithm to select individuals with large fitness function values ​​from the current population to enter the next generation; A6: Perform crossover and mutation operations to generate a new generation of population; A7: Determine whether the current iteration meets the preset termination condition. If so, the process ends; otherwise, the process returns to A5. A8: Select the individual with the largest fitness function value as the optimized parameter, and set it as the control parameter of the vibrating screen to be tested.

10. The screening efficiency optimization method for a vibrating screen according to claim 8, characterized in that: The fitness evaluation function F(x) is: F(x)=w1y1-w2y2-w3y3, Among them, y1, y2, and y3 represent screening efficiency, energy consumption, and production cost respectively, and w1, w2, and w3 are weight coefficients.

Images