Precise control method and system for centrifuge parameters used in pipe pile centrifugal process

Abstract

The present invention discloses a method and system for accurately controlling centrifuge parameters for a pipe pile centrifugal process, and the present invention relates to the technical field of pipe pile production process. The method comprises the following steps: the centrifuge parameters in historical production and the strength data of the produced pipe piles, the centrifuge parameters including the centrifuge speed, the centrifugal time and the vibration frequency parameters, generate a training sample set; establish a pipe pile strength prediction model, use the data in the training sample set as input, and use the pipe pile strength as a label to train the prediction model; input the real-time centrifuge parameters into the model, if the production prediction strength data is greater than or equal to the required strength, keep the current centrifuge parameters, if the production prediction strength data is less than the required strength, correct the centrifuge parameters based on the density and resonance frequency of the pipe pile, obtain the corrected parameters of the centrifuge, output the corrected parameters of the centrifuge to accurately control the centrifuge.

Description

Technical Field

[0001] The present invention relates to the technical field of pipe pile production technology, and in particular to a method and system for accurately controlling centrifuge parameters used in a pipe pile centrifugal process. Background Art

[0002] In modern construction and infrastructure engineering, pipe piles are an important basic support structure and are widely used in the construction of various buildings, bridges and other infrastructure. The strength and density of pipe piles are directly related to their bearing capacity and durability. Therefore, during the production process, their strength needs to be strictly controlled and predicted. Traditional pipe pile production processes rely more on experience and manual monitoring, and lack systematic parameter control, resulting in instability in the strength of the produced pipe piles, making it difficult to meet the increasingly stringent engineering standards and safety requirements.

[0003] Especially in the centrifugal process, centrifuge parameters such as rotation speed, centrifugal time and vibration frequency have an important influence on the strength and density of the final product. However, the specific parameter settings often lack scientific basis, which can easily lead to problems of insufficient or excessive strength. With the development of building materials technology, the market's requirements for the strength of pipe piles are constantly increasing, which makes traditional production methods increasingly unsuitable. Therefore, a new method is needed to achieve precise control of centrifuge parameters to ensure that the strength of the pipe piles meets the design standards.

[0004] In the prior art, the publication number CN114536541A discloses a pipe pile production method, equipment and storage medium based on the pipe pile centrifugal process, wherein the method includes real-time acquisition of amplitude data at the rotating shaft of the centrifuge during the centrifugal process, determining the deflection change characteristics of the rotating shaft according to the amplitude data; judging whether the deflection change of the rotating shaft tends to be constant based on the deflection change characteristics, and sending a stop operation instruction or a speed adjustment instruction to the centrifuge when the deflection change is constant. This scheme only relies on the deflection change characteristics of the rotating shaft to judge the operating state, which may lead to insufficient sensitivity to vibration and resonance. The scheme focuses on the deflection change of the rotating shaft, while ignoring the monitoring of other key parameters (such as rotation speed, centrifugal time, resonance frequency, etc.). These factors may have an important impact on the strength and density of the pipe pile. A single reliance on deflection changes may not fully reflect the production status, resulting in reduced applicability and accuracy of the system, making the quality of the produced pipe piles unguaranteed, and the controllability of the production process is weak.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The object of the present invention is to provide a method and system for accurately controlling centrifuge parameters for a pipe pile centrifugal process, so as to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A method for accurately controlling centrifuge parameters for a pipe pile centrifugal process, the specific steps comprising:

[0009] S1, obtaining several groups of centrifuge parameters and strength data of produced pipe piles in historical production, preprocessing the centrifuge parameters and corresponding strength data, and generating a training sample set, wherein the centrifuge parameters include rotation speed, centrifugation time and vibration frequency;

[0010] S2, establishing a pile strength prediction model, taking the centrifuge parameters in the training sample set as input data, and taking the corresponding pile strength data as labels to train the prediction model, to obtain a pile strength prediction model with centrifuge parameters as input and pile strength data as output;

[0011] S3, input the real-time centrifuge parameters into the pipe pile strength prediction model to obtain the production prediction strength data of the pipe pile, compare the production prediction strength data with the demand strength, if the production prediction strength data is greater than or equal to the demand strength, keep the current centrifuge parameters, if the production prediction strength data is less than the demand strength, execute step S4;

[0012] S4, based on the density of the pipe pile and the resonance frequency of the pipe pile, the corresponding centrifuge parameters are corrected to obtain the corrected parameters of the centrifuge, and the corrected parameters of the centrifuge are output to accurately control the centrifuge, wherein the corrected parameters of the centrifuge include a speed correction value, a centrifugal time correction value and a vibration frequency correction value.

[0013] Furthermore, the centrifuge parameters and the corresponding intensity data are preprocessed, and the preprocessing includes data standardization and data cleaning, wherein the formula for data standardization is:

[0014]

[0015] Where N i is the standardized i-th speed data, n i Indicates the i-th speed data collected, μ n is the average value of the speed data, σ n is the variance of the speed data;

[0016] Data cleaning is performed based on the standardized data. The logic of data cleaning is as follows: setting an abnormal threshold Y,

[0017]

[0018] Where n icl represents the speed data after the i-th cleaning, |N i | represents the absolute value of the standardized i-th speed data. The centrifugal time, vibration frequency and pipe pile strength data are preprocessed according to the same method to obtain the preprocessed centrifugal time, vibration frequency and pipe pile strength data.

[0019] Furthermore, a pile strength prediction model is established based on a long short-term memory network model, and the hyperparameters of the LSTM model are set at the same time. The hyperparameters of the LSTM model include: the number of network layers, the number of iterations, the learning rate, the batch size, the number of training times, the batch processing number, and the number of hidden layer neurons;

[0020] The number of network layers is set to a 3-layer network structure, the number of iterations is set to 100, the learning rate is set to 0.001, the batch size is set to 32, the number of training times is set to 100, and the number of hidden layer neurons is 64. After the training is completed, a pipe pile strength prediction model is obtained, whose input includes centrifuge speed, centrifugation time and vibration frequency, and the output is the predicted value of pipe pile strength.

[0021] Furthermore, the centrifuge parameters collected in real time include the real-time speed N s , Real-time centrifugation time T s and real-time vibration frequency f s After preprocessing, the data is input into the trained pile strength prediction model to obtain the predicted value of the pile strength under the current parameters. The mean and variance used for data standardization follow the historical data, and the calibrated pile strength prediction value is Q for , calibrate the demand intensity to Q tar ;

[0022] When Q for ≥Q tar When the real-time speed N is output s , Real-time centrifugation time T s and real-time vibration frequency f s to carry out work;

[0023] When Q for tar When the real-time speed N s , Real-time centrifugation time T s and real-time vibration frequency f s Make corrections.

[0024] Furthermore, the step S4 comprises the following steps:

[0025] S41, based on the density and strength error of the pile, the real-time rotation speed N s ​Make corrections to obtain the speed correction value. The speed correction value is calculated based on the formula:

[0026] N corr =N s +k1*(ρ p -ρ tar )-ω*(Q tar -Q for )

[0027] Where N corr represents the rotation speed correction value, k1 is the weight coefficient of the pile density error, ρ p Indicates the density of the pile, ρ tar Indicates the reference density, (Q tar -Q for ) is the intensity error, ω represents the weight coefficient of the intensity error, where ω and k1 are both greater than 0;

[0028] S42, based on the intensity error, the real-time centrifugation time T s Correction is performed to obtain the centrifugal time correction value, and the formula for calculating the centrifugal time correction value is:

[0029] T corr =T s +ω*(Q tar -Q for )

[0030] Where, T corr Indicates the centrifugal time correction value;

[0031] S43, based on the resonance frequency and strength error of the pile, the real-time vibration frequency f s Correction is performed to obtain the vibration frequency correction value. The vibration frequency correction value is calculated based on the formula:

[0032]

[0033] In the formula, f corr is the vibration frequency correction value, k2 represents the safety factor, f r Represents the resonant frequency of the pile, where k2 is greater than 0.

[0034] Furthermore, the step S41 includes the following steps:

[0035] S411, the formula for calculating the compactness of the pile based on the initial density of the pile material and the centrifugal force is:

[0036]

[0037] In the formula, ρ0 is the initial density of the pile material, l is the compression coefficient of the material, F represents the centrifugal force, and g represents the acceleration of gravity;

[0038] The formula for calculating the centrifugal force F is:

[0039]

[0040] In the formula, m is the material mass, and R represents the centrifugal radius;

[0041] At the same time, step S43 includes the following steps:

[0042] S431, the formula for calculating the resonance frequency of the pipe pile is:

[0043]

[0044] In the formula, G represents the stiffness of the pile;

[0045] The formula for calculating the stiffness G of the pipe pile is:

[0046]

[0047] Where E is the Young's modulus of the pile material, I is the moment of inertia of the pile section, and L is the length of the pile.

[0048] The present invention also provides a precise control system for centrifuge parameters used in a pipe pile centrifugal process, wherein the precise control system for centrifuge parameters used in a pipe pile centrifugal process is used to execute the precise control method for centrifuge parameters used in a pipe pile centrifugal process, comprising:

[0049] A historical data preprocessing module is used to obtain several groups of centrifuge parameters and strength data of produced pipe piles in historical production, preprocess the centrifuge parameters and corresponding strength data, and generate a training sample set. The centrifuge parameters include rotation speed, centrifugal time and vibration frequency;

[0050] The prediction model building module is used to build a pile strength prediction model, taking the centrifuge parameters in the training sample set as input data and using the corresponding pile strength data as labels to train the prediction model, thereby obtaining a pile strength prediction model with centrifuge parameters as input and pile strength data as output;

[0051] A real-time parameter control module is used to input the real-time centrifuge parameters into the pipe pile strength prediction model to obtain the production prediction strength data of the pipe pile, compare the production prediction strength data with the demand strength, and if the production prediction strength data is greater than or equal to the demand strength, keep the current centrifuge parameters; if the production prediction strength data is less than the demand strength, input the current centrifuge parameters into the centrifuge parameter correction module for correction;

[0052] The centrifuge parameter correction module is used to correct the current corresponding centrifuge parameters based on the density of the pipe pile and the resonance frequency of the pipe pile to obtain the correction parameters of the centrifuge, and output the correction parameters of the centrifuge to accurately control the centrifuge. The correction parameters of the centrifuge include a speed correction value, a centrifugal time correction value and a vibration frequency correction value.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] The strength data of several groups of historically produced pipe piles and the centrifuge parameters corresponding to the strength data are obtained, and the centrifuge parameters and the corresponding strength data are preprocessed to produce a training sample set, wherein the centrifuge parameters include rotation speed, centrifugal time and vibration frequency; a foundation is established for establishing a strength prediction model for multiple centrifuge parameters, which improves the diversity of data and the accuracy of prediction, and establishes a pipe pile strength prediction model, which can realize the precise control of centrifuge parameters, thereby effectively solving the problem of substandard strength of pipe piles. Through the analysis and processing of historical production data, a data-driven strength prediction system is formed, which enables the centrifuge parameters in the production process to be dynamically adjusted, thereby ensuring the quality and safety of pipe piles. At the same time, the corresponding centrifuge parameters are corrected based on the density of the pipe pile and the resonance frequency of the pipe pile, which not only improves the automation and intelligent level of pipe pile production, but also greatly reduces the scrap rate caused by improper parameter setting. Thereby reducing production costs and improving resource utilization efficiency, at the same time, the introduction of real-time monitoring and feedback mechanisms enhances the controllability of the production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0056] Figure 2 It is a schematic diagram of the overall system structure of the present invention. DETAILED DESCRIPTION

[0057] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0058] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0059] Example:

[0060] See also Figure 1 , the present invention provides a technical solution:

[0061] A method for accurately controlling centrifuge parameters for a pipe pile centrifugal process, the specific steps comprising:

[0062] S1, obtaining several groups of centrifuge parameters and strength data of produced pipe piles in historical production, preprocessing the centrifuge parameters and corresponding strength data, and generating a training sample set, wherein the centrifuge parameters include rotation speed, centrifugation time and vibration frequency;

[0063] The centrifuge parameters and the corresponding intensity data are preprocessed, and the preprocessing includes data standardization and data cleaning, wherein the formula for data standardization is:

[0064]

[0065] Where N i is the standardized i-th speed data, n i Indicates the i-th speed data collected, μ n is the average value of the speed data, σ n is the variance of the speed data;

[0066] Data cleaning is performed based on the standardized data. The logic of data cleaning is as follows: setting an abnormal threshold Y,

[0067]

[0068] Where n icl represents the speed data after the i-th cleaning, |N i | represents the absolute value of the standardized i-th speed data; |N i|Data greater than the abnormal threshold Y is marked as abnormal data, where the abnormal threshold Y is generally taken as 3; the abnormal speed data is cleared, and the corresponding centrifugal time, vibration frequency and strength data are cleared at the same time. The preprocessing method for centrifugal time and vibration frequency is the same as that for speed. The centrifugal time and vibration frequency are preprocessed in the same way to obtain the preprocessed centrifugal time, vibration frequency and pile strength data. A training sample set is generated based on the preprocessed data, including speed, centrifugal time and vibration frequency and the corresponding pile strength data. The pile strength data is compressive strength, which refers to the maximum pressure that concrete can withstand under pressure, and is an important indicator for evaluating concrete quality and structural safety.

[0069] S2, establishing a pile strength prediction model, taking the centrifuge parameters in the training sample set as input data, and taking the corresponding pile strength data as labels to train the prediction model, to obtain a pile strength prediction model with centrifuge parameters as input and pile strength data as output;

[0070] A pipe pile strength prediction model is established based on a long short-term memory network model, the long short-term memory network model is an LSTM model, and an activation function and an optimization algorithm are selected, wherein the Tanh function is selected as the activation function, and Adam is selected as the optimization algorithm of the LSTM model; the formula of the Tanh function is:

[0071]

[0072] In the formula, f(x) represents the Tanh function, and the independent variable r represents the weighted sum of the neuron's input, that is, the result of the weighted summation of the input received by the neuron from the previous layer;

[0073] At the same time, the hyperparameters of the LSTM model are set, and the hyperparameters of the LSTM model include: the number of network layers, the number of iterations, the learning rate, the batch size, the number of training times, the batch processing number, and the number of hidden layer neurons;

[0074] The number of network layers is set to a 3-layer network structure, the number of iterations is set to 100, the learning rate is set to 0.001, the batch size is set to 32, the number of training times is set to 100, and the number of hidden layer neurons is 64. After the training is completed, a pipe pile strength prediction model is obtained, whose input includes centrifuge speed, centrifugation time and vibration frequency, and the output is the predicted value of pipe pile strength.

[0075] S3, input the real-time centrifuge parameters into the pipe pile strength prediction model to obtain the production prediction strength data of the pipe pile, compare the production prediction strength data with the demand strength, if the production prediction strength data is greater than or equal to the demand strength, keep the current centrifuge parameters, if the production prediction strength data is less than the demand strength, execute step S4;

[0076] The real-time collected centrifuge parameters include the real-time speed N s , Real-time centrifugation time T s and real-time vibration frequency f s After preprocessing, the data is input into the trained pile strength prediction model. The mean and variance used for data standardization follow the historical data to obtain the predicted value of the pile strength under the current parameters. The calibrated predicted value of the pile strength is Q for , calibrate the demand intensity to Q tar ;

[0077] When Q for ≥Q tar When the real-time speed N is output s , Real-time centrifugation time T s and real-time vibration frequency f s , indicating that under the current parameters, the produced pipe piles can meet the actual work requirements, and no adjustment is needed, and the current parameters can be maintained for production;

[0078] When Q for tar When the current parameters are used, the strength of the produced piles does not meet the actual required strength, and the current real-time speed N needs to be adjusted. s , Real-time centrifugation time T s and real-time vibration frequency f s Make modifications to meet actual needs.

[0079] S4, based on the density of the pipe pile and the resonance frequency of the pipe pile, the corresponding centrifuge parameters are corrected to obtain the corrected parameters of the centrifuge, and the corrected parameters of the centrifuge are output to accurately control the centrifuge, wherein the corrected parameters of the centrifuge include a speed correction value, a centrifugal time correction value and a vibration frequency correction value.

[0080] The step S4 comprises the following steps:

[0081] S41, based on the density and strength error of the pile, the real-time rotation speed N s Make corrections to obtain the speed correction value. The speed correction value is calculated based on the formula:

[0082] N corr =N s +k1*(ρ p -ρ tar )-ω*(Q tar -Q for )

[0083] Where N corr represents the rotation speed correction value, k1 is the weight coefficient of the pile density error, ρ p Indicates the density of the pile, ρ​tar Indicates the reference density, (Q tar -Q for ) is the strength error, ω represents the weight coefficient of the strength error, where ω and k1 are both greater than 0; and since the strength error has a greater impact on the speed correction value, ω>k1, where k1*(ρ p -ρ tar ) represents the adjustment of the rotation speed according to the deviation of the pile density; when ρ p >ρ tar When ρ, the system density is high, which means that the speed needs to be appropriately reduced to reduce the risk. p < tar When the system density is low, the speed can be appropriately increased.

[0084] When the strength of the produced pile does not reach the expected strength, the speed of the centrifuge should usually be reduced. The higher the speed of the centrifuge, the greater the centrifugal force generated, which may cause the separation of particles inside the concrete, affecting the uniformity and density of the material, thereby reducing the strength of the pile. Therefore, the speed correction value N corr Inversely proportional to the intensity error.

[0085] S42, based on the intensity error, the real-time centrifugation time T s Correction is performed to obtain the centrifugal time correction value, and the formula for calculating the centrifugal time correction value is:

[0086] T corr =T s +ω*(Q tar -Q for )

[0087] Where, T corr Indicates the centrifugal time correction value;

[0088] When the strength of the produced pipe piles does not reach the expected strength, the centrifugal time should usually be increased. Increasing the centrifugal time can make the concrete more fully compacted under the action of centrifugal force, help reduce pores, increase density, and thus improve the final strength. Therefore, the centrifugal time correction value is proportional to the strength error.

[0089] S43, based on the resonance frequency and strength error of the pile, the real-time vibration frequency f s Correction is performed to obtain the vibration frequency correction value. The vibration frequency correction value is calculated based on the formula:

[0090]

[0091] In the formula, f corr is the vibration frequency correction value, k2 represents the safety factor, f rRepresents the resonant frequency of the pile, where k2 is greater than 0, and because it is necessary to avoid the vibration frequency approaching the resonant frequency, k2>ω;

[0092] When the strength of the produced pile does not reach the expected strength, the vibration frequency should usually be reduced and the vibration frequency of the centrifuge should be adjusted to keep it away from the resonance frequency to reduce the risk of resonance. Therefore, the vibration frequency correction value is related to the real-time vibration frequency f s and the resonant frequency f of the pile r The closer the real-time vibration frequency is to the resonance frequency, the more the vibration frequency is adjusted to be reduced. The safety factor k2 can be calculated by the ratio of the maximum bearing capacity of the pile to the maximum stress under dynamic response.

[0093] The step S41 comprises the following steps:

[0094] S411, the formula for calculating the compactness of the pile based on the initial density of the pile material and the centrifugal force is:

[0095]

[0096] In the formula, ρ0 is the initial density of the pile material, l is the compression coefficient of the material, F represents the centrifugal force, and g represents the acceleration of gravity, which is usually taken as 9.8m / s 2 ;

[0097] The formula for calculating the centrifugal force F is:

[0098]

[0099] In the formula, m is the material mass, and R represents the centrifugal radius;

[0100] At the same time, step S43 includes the following steps:

[0101] S431, the formula for calculating the resonance frequency of the pipe pile is:

[0102]

[0103] In the formula, G represents the stiffness of the pile;

[0104] The formula for calculating the stiffness G of the pipe pile is:

[0105]

[0106] Where E is the Young's modulus of the pile material, I is the moment of inertia of the pile section, and L is the length of the pile.

[0107] Among them, Young's modulus (elastic modulus) is an inherent property of the material, which reflects the ratio of strain to stress of the material when it is subjected to force. Young's modulus can usually be obtained by the following methods: obtain the stress-strain curve through standard compression test or tensile test, and calculate the Young's modulus. For concrete, cubic or cylindrical specimens are usually used for testing. For common concrete, the empirical formula of Young's modulus can be used for estimation, which is usually between 25-30GPa (GigaPascals). The specific value will vary according to the strength grade and mix ratio of the concrete.

[0108] For hollow pipe piles, the moment of inertia can be calculated based on the outer diameter and inner diameter, based on the formula:

[0109]

[0110] Where D out Indicates the outer diameter of the pile, D inn Indicates the inner diameter of the pile.

[0111] See also Figure 2 ,

[0112] The present invention also provides a precise control system for centrifuge parameters used in a pipe pile centrifugal process, wherein the precise control system for centrifuge parameters used in a pipe pile centrifugal process is used to execute the precise control method for centrifuge parameters used in a pipe pile centrifugal process, comprising:

[0113] A historical data preprocessing module is used to obtain several groups of centrifuge parameters and strength data of produced pipe piles in historical production, preprocess the centrifuge parameters and corresponding strength data, and generate a training sample set. The centrifuge parameters include rotation speed, centrifugal time and vibration frequency;

[0114] The prediction model building module is used to build a pile strength prediction model, taking the centrifuge parameters in the training sample set as input data and using the corresponding pile strength data as labels to train the prediction model, thereby obtaining a pile strength prediction model with centrifuge parameters as input and pile strength data as output;

[0115] A real-time parameter control module is used to input the real-time centrifuge parameters into the pipe pile strength prediction model to obtain the production prediction strength data of the pipe pile, compare the production prediction strength data with the demand strength, and if the production prediction strength data is greater than or equal to the demand strength, keep the current centrifuge parameters; if the production prediction strength data is less than the demand strength, input the current centrifuge parameters into the centrifuge parameter correction module for correction;

[0116] The centrifuge parameter correction module is used to correct the current corresponding centrifuge parameters based on the density of the pipe pile and the resonance frequency of the pipe pile to obtain the correction parameters of the centrifuge, and output the correction parameters of the centrifuge to accurately control the centrifuge. The correction parameters of the centrifuge include a speed correction value, a centrifugal time correction value and a vibration frequency correction value.

[0117] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0118] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for accurately controlling centrifuge parameters for a pipe pile centrifugal process, characterized in that: The specific steps include: S1, obtaining several groups of centrifuge parameters and strength data of produced pipe piles in historical production, preprocessing the centrifuge parameters and corresponding strength data, and generating a training sample set, wherein the centrifuge parameters include rotation speed, centrifugation time and vibration frequency; S2, establishing a pile strength prediction model, taking the centrifuge parameters in the training sample set as input data, and taking the corresponding pile strength data as labels to train the prediction model, to obtain a pile strength prediction model with centrifuge parameters as input and pile strength data as output; S3, input the real-time centrifuge parameters into the pipe pile strength prediction model to obtain the production prediction strength data of the pipe pile, compare the production prediction strength data with the demand strength, if the production prediction strength data is greater than or equal to the demand strength, keep the current centrifuge parameters, if the production prediction strength data is less than the demand strength, execute step S4; The real-time collected centrifuge parameters include the real-time speed N s , Real-time centrifugation time T s and real-time vibration frequency f s After preprocessing, the data is input into the trained pile strength prediction model to obtain the predicted value of the pile strength under the current parameters. The mean and variance used for data standardization follow the historical data, and the calibrated pile strength prediction value is Q for , calibrate the demand intensity to Q tar ; When Q for ≥Q tar When the real-time speed N is output s , Real-time centrifugation time T s and real-time vibration frequency f s to carry out work; When Q for tar When the real-time speed N s , Real-time centrifugation time T s and real-time vibration frequency f s make corrections;​ S4, based on the density of the pipe pile and the resonance frequency of the pipe pile, the corresponding centrifuge parameters are corrected to obtain the corrected parameters of the centrifuge, and the corrected parameters of the centrifuge are output to accurately control the centrifuge, wherein the corrected parameters of the centrifuge include a speed correction value, a centrifugal time correction value, and a vibration frequency correction value; The step S4 comprises the following steps: S41, based on the density and strength error of the pile, the real-time rotation speed N s Make corrections to obtain the speed correction value. The speed correction value is calculated based on the formula: N corr =N s +k1*(ρ p -r tar )-ω*(Q tar -Q for ) Where N corr represents the rotation speed correction value, k1 is the weight coefficient of the pile density error, ρ p Indicates the density of the pile, ρ tar Indicates the reference density, (Q tar -Q for ) is the intensity error, ω represents the weight coefficient of the intensity error, where ω and k1 are both greater than 0; S42, based on the intensity error, the real-time centrifugation time T s Correction is performed to obtain the centrifugal time correction value, and the formula for calculating the centrifugal time correction value is: T corr =T s +ω*(Q tar -Q for ) Where, T corr Indicates the centrifugal time correction value; S43, based on the resonance frequency and strength error of the pile, the real-time vibration frequency f s Correction is performed to obtain the vibration frequency correction value. The vibration frequency correction value is calculated based on the formula: In the formula, f corr is the vibration frequency correction value, k2 represents the safety factor, f r Represents the resonant frequency of the pipe pile, where k2 is greater than 0; The step S41 comprises the following steps: S411, the formula for calculating the compactness of the pile based on the initial density of the pile material and the centrifugal force is: In the formula, ρ0 is the initial density of the pile material, l is the compression coefficient of the material, F represents the centrifugal force, and g represents the acceleration of gravity; The formula for calculating the centrifugal force F is: In the formula, m is the material mass, and R represents the centrifugal radius; At the same time, step S43 includes the following steps: S431, the formula for calculating the resonance frequency of the pipe pile is: In the formula, G represents the stiffness of the pile; The formula for calculating the stiffness G of the pipe pile is: Where E is the Young's modulus of the pile material, I is the moment of inertia of the pile section, and L is the length of the pile.

2. The method for accurately controlling centrifuge parameters for a pipe pile centrifugal process according to claim 1, characterized in that: The centrifuge parameters and the corresponding intensity data are preprocessed, and the preprocessing includes data standardization and data cleaning, wherein the formula for data standardization is: Where N i is the standardized i-th speed data, n i Indicates the i-th speed data collected, μ n is the average value of the speed data, σ n is the variance of the speed data; Data cleaning is performed based on the standardized data. The logic of data cleaning is as follows: setting an abnormal threshold Y, Where n icl represents the speed data after the i-th cleaning, |N i | represents the absolute value of the standardized i-th speed data. The centrifugal time, vibration frequency and pipe pile strength data are preprocessed according to the same method to obtain the preprocessed centrifugal time, vibration frequency and pipe pile strength data.

3. The method for accurately controlling centrifuge parameters for a pipe pile centrifugal process according to claim 1, characterized in that: A pile strength prediction model is established based on a long short-term memory network model, and the hyperparameters of the LSTM model are set. The hyperparameters of the LSTM model include: the number of network layers, the number of iterations, the learning rate, the batch size, the number of training times, the batch processing number, and the number of hidden layer neurons; The number of network layers is set to a 3-layer network structure, the number of iterations is set to 100, the learning rate is set to 0.001, the batch size is set to 32, the number of training times is set to 100, and the number of hidden layer neurons is 64. After the training is completed, a pipe pile strength prediction model is obtained, whose input includes centrifuge speed, centrifugation time and vibration frequency, and the output is the predicted value of pipe pile strength.

4. A precise control system for centrifuge parameters used in a pipe pile centrifugal process, characterized in that: The precise control system for centrifuge parameters used in the pipe pile centrifugal process is used to execute the precise control method for centrifuge parameters used in the pipe pile centrifugal process according to any one of claims 1 to 3, comprising: A historical data preprocessing module is used to obtain several groups of centrifuge parameters and strength data of produced pipe piles in historical production, preprocess the centrifuge parameters and corresponding strength data, and generate a training sample set. The centrifuge parameters include rotation speed, centrifugal time and vibration frequency; The prediction model building module is used to build a pile strength prediction model, taking the centrifuge parameters in the training sample set as input data and using the corresponding pile strength data as labels to train the prediction model, thereby obtaining a pile strength prediction model with centrifuge parameters as input and pile strength data as output; A real-time parameter control module is used to input the real-time centrifuge parameters into the pipe pile strength prediction model to obtain the production prediction strength data of the pipe pile, compare the production prediction strength data with the demand strength, and if the production prediction strength data is greater than or equal to the demand strength, keep the current centrifuge parameters; if the production prediction strength data is less than the demand strength, input the current centrifuge parameters into the centrifuge parameter correction module for correction; The centrifuge parameter correction module is used to correct the current corresponding centrifuge parameters based on the density of the pipe pile and the resonance frequency of the pipe pile to obtain the correction parameters of the centrifuge, and output the correction parameters of the centrifuge to accurately control the centrifuge. The correction parameters of the centrifuge include a speed correction value, a centrifugal time correction value and a vibration frequency correction value.

Images