Rib winding machine parameter determination method, medium and system

Abstract

The invention provides a rib winding machine parameter determination method, medium and system, and belongs to the technical field of rib winding machines, and the method comprises the steps: collecting data: rib winding parameters (speed, tension, angle, spacing, number of turns and pressure) and rib winding quality (uniformity, strength, smoothness, compactness and precision); the parameters comprise the diameter, the strength grade, the surface state, the elastic modulus, the yield strength and the ductility; establishing equations including a uniformity equation, a strength equation, a smoothness equation, a compactness equation and the like; solving the equation set to obtain an analytical solution of each process parameter; establishing a multi-objective optimization model, and solving optimized process parameters by taking the analytical solution weighted sum as an objective; and an optimization result is evaluated through a test, and the optimal parameter is selected as an optimal parameter. By means of mathematical modeling and multi-objective optimization, the complex relation between the rib winding process parameters and the rib winding quality is systematically analyzed and solved, and the optimal parameter combination meeting multiple quality index requirements is obtained.

Description

Technical Field

[0001] The present invention belongs to the technical field of rib winding machines, and in particular, relates to a method, medium and system for determining parameters of a rib winding machine. Background Art

[0002] Reinforced concrete structure is an important structural form widely used in modern buildings. It has excellent performance in terms of bearing capacity, earthquake resistance, durability, etc., and has been widely used in residential, highway, bridge and other fields. In the manufacturing process of reinforced concrete structure, rebar winding is a key process, which directly affects the quality and service life of the entire structure. The quality of the rebar winding process not only determines the force performance of the steel bar, but also affects the bonding strength between concrete and steel bars, thereby affecting the overall performance of the entire structure. Therefore, how to determine the optimal process parameters of the rebar winding machine has always been an important problem that needs to be solved in the field of construction.

[0003] At present, the determination of process parameters of rib winding machines mainly depends on experience accumulation and repeated experiments, and it is difficult to obtain the optimal combination of process parameters. Summary of the invention

[0004] In view of this, the present invention provides a method, medium and system for determining parameters of a rib winding machine, which can solve the technical problem that the current determination of process parameters of a rib winding machine mainly relies on experience accumulation and repeated experiments, and it is difficult to obtain the optimal process parameter combination.

[0005] The present invention is achieved in that:

[0006] A first aspect of the present invention provides a method for determining parameters of a tendon winding machine, comprising the following steps:

[0007] S10, performing multiple groups of rebar winding tests, collecting rebar winding parameters and rebar winding quality under different process parameters and different rebar material parameters, wherein the rebar winding parameters at least include rebar winding speed, rebar winding tension, rebar winding angle, rebar winding spacing, rebar winding number of turns, and rebar winding pressure, and the rebar winding quality at least includes rebar winding uniformity, rebar winding strength, rebar winding surface finish, rebar winding tightness, and rebar winding position accuracy; the rebar material parameters at least include rebar diameter, rebar strength grade, rebar surface state, rebar elastic modulus, rebar yield strength, and rebar elongation;

[0008] S20, establishing a group of equations for the quality of the winding process, including a winding uniformity equation, a winding strength equation, a winding surface finish equation, and a winding tightness equation;

[0009] S30, solving the process quality equation group to obtain multiple analytical solutions, which are respectively an analytical solution for rebar winding speed, an analytical solution for rebar winding tension, an analytical solution for rebar winding angle, an analytical solution for rebar winding spacing, an analytical solution for rebar winding number of turns, and an analytical solution for rebar winding pressure;

[0010] S40, establishing a multi-objective optimization model by taking the weighted sum of multiple analytical solutions as the objective function and taking the boundary conditions of the rib winding machine process and the material performance limitations as the constraint conditions;

[0011] S50, solving the multi-objective optimization model to obtain multiple groups of optimized process parameters;

[0012] S60, using the optimized process parameters to perform actual rib winding tests, and performing quality evaluation on the rib winding results, and recording the evaluation results of the rib winding quality indicators corresponding to each set of parameters;

[0013] S70. According to the evaluation result of the rib winding quality index, the best performing process parameters are selected as the optimal rib winding machine parameters and outputted.

[0014] Wherein, the step S10 specifically includes: collecting multiple sets of rebar winding test data, recording rebar winding parameters and rebar winding quality indicators under different process parameters and rebar material parameter conditions. The rebar winding parameters at least include rebar winding speed, rebar winding tension, rebar winding angle, rebar winding spacing, rebar winding number of turns and rebar winding pressure, etc.; the rebar winding quality indicators at least include rebar winding uniformity, rebar winding strength, rebar winding surface finish, rebar winding tightness and rebar winding position accuracy, etc. The rebar material parameters at least include rebar diameter, rebar strength grade, rebar surface state, rebar elastic modulus, rebar yield strength and rebar elongation, etc.

[0015] Wherein, the step S20 specifically includes: according to the data collected in step S10, establishing a set of equations for the quality of the winding process, including a winding uniformity equation, a winding strength equation, a winding surface finish equation, and a winding tightness equation, etc., to reflect the relationship between the winding process parameters and the winding quality indicators. The set of equations can be fitted using a data-driven modeling method such as multiple linear regression and artificial neural network, and its fitting accuracy should meet certain requirements, such as R 2 Greater than 0.9.

[0016] Wherein, the step S30 specifically includes: solving the set of equations for the quality of the winding process established in step S20, and obtaining analytical solutions for the winding speed, winding tension, winding angle, winding spacing, winding turns, and winding pressure, etc. The solution process should meet the actual process boundary conditions and material performance limitations of the winding machine, and the results should be accurate and stable.

[0017] Wherein, the step S40 specifically includes: establishing a multi-objective optimization model, taking the weighted sum of the analytical solutions of the rebar winding process parameters solved in step S30 as the objective function, and taking the process boundary conditions of the rebar winding machine and the performance restrictions of the steel bar material as constraints. The optimization model can be solved by using a meta-heuristic optimization algorithm such as a genetic algorithm and a particle swarm optimization.

[0018] Wherein, the step S50 specifically includes: solving the multi-objective optimization model established in step S40 to obtain multiple sets of optimized rib winding process parameters. The selection of the optimization algorithm should be determined according to the characteristics of the problem, such as weighted sum method, NSGA-II, etc., and ensure that the optimization result converges stably and the Pareto front solution set is relatively balanced.

[0019] Wherein, the step S60 specifically includes: using the optimized parameters obtained in step S50 to perform an actual rib winding test, measuring and recording rib winding uniformity, rib winding strength, rib winding surface finish, rib winding tightness, rib winding position accuracy and other rib winding quality indicators. The test conditions should be the same as those in step S10 to ensure comparability of the results.

[0020] Wherein, the step S70 specifically includes: according to the test results of step S60, selecting the winding process parameters with the best comprehensive performance as the final optimal parameters. When selecting, corresponding threshold requirements can be set for each winding quality index according to actual needs, and the parameter combination with the best comprehensive performance can be selected on the premise of meeting the threshold.

[0021] Optionally, the selected parameter combination should be able to stably meet the quality requirements of the rib winding in actual production, while taking into account the operability and economy of the process.

[0022] The uniformity equation of the reinforcement winding is specifically:

[0023] U=f U (v,T,θ,d);

[0024] The polynomial regression method is used, which is specifically expressed as:

[0025] U=a0+a1v+a2T+a3θ+a4d+a5v 2 +a6T 2 +a7θ 2 +a8d 2 +a9vT+a 10 vθ+a 11 vd+a 12 Tθ+a 13 Td+a 14 θd;

[0026] Where a0, a1, …, a 14 is the unknown coefficient, v is the winding speed, T is the winding tension, θ is the winding angle, and d is the winding spacing.

[0027] The strength equation of the reinforcing bar is specifically:

[0028] S=f S (v,T,θ,d,n,p,D,f,s,E,σy ,ε);

[0029] Where S is the strength of the steel bar, v is the speed of the steel bar, n is the number of turns, p is the pressure of the steel bar, D is the diameter of the steel bar, f is the strength grade of the steel bar, s is the surface state of the steel bar, E is the elastic modulus of the steel bar, σ y is the yield strength of the steel bar, and ε is the elongation of the steel bar. The above formula uses polynomial regression. Considering that there are many variables, the principal component analysis method is first used to reduce the dimension, and then the following regression equation is established:

[0030]

[0031] Where PC1, PC2, …, PC m is the principal component, m is the number of principal components selected, and b0, b1,… are the coefficients to be determined.

[0032] The surface finish equation of the rib winding is specifically:

[0033] R = f R (v,T,p,s);

[0034] Considering the nonlinear relationship between surface finish and the winding speed v, winding tension T, winding pressure p, and steel bar surface state s, a smooth function is used for fitting, as follows:

[0035] R=c0+f1(v)+f2(T)+f3(p)+f4(s);

[0036] Where f1(v), f2(T), f3(p), and f4(s) are smooth functions, which respectively describe the nonlinear effects of winding speed, winding tension, winding pressure, and steel bar surface state on surface finish, and are fitted using spline functions.

[0037] The winding tightness equation is specifically:

[0038] C = f C (T, p, D, E);

[0039] Considering the exponential relationship between tightness, tension and pressure, a nonlinear regression model is used, as follows:

[0040]

[0041] In the formula, d0, d1,…, d6 are unknown coefficients.

[0042] The analytical solution of the reinforcing bar winding speed is specifically:

[0043]

[0044] In the formula, v* is the optimal winding speed, v0 is the initial winding speed, and T0 is the initial winding tension;

[0045] The analytical solution of the reinforcing bar tension is specifically:

[0046]

[0047] Where, T * is the optimal tendon tension;

[0048] The analytical solution of the reinforcing bar angle is specifically:

[0049]

[0050] In the formula, θ * is the optimal reinforcement winding angle, θ0 is the initial reinforcement winding angle, and n0 is the initial number of reinforcement winding turns;

[0051] The analytical solution for the spacing between reinforcing bars is specifically:

[0052]

[0053] Where, d * is the optimal spacing of reinforcing bars, d0 is the initial spacing of reinforcing bars;

[0054] The analytical solution for the number of reinforcing bar turns is specifically:

[0055]

[0056] Where n * is the optimal number of reinforcing turns, n0 is the initial number of reinforcing turns;

[0057] The analytical solution of the reinforcing bar pressure is specifically:

[0058]

[0059] In the formula, p * is the optimal reinforcement pressure, and p0 is the initial reinforcement pressure.

[0060] The objective function of the multi-objective optimization model is specifically:

[0061] F=w1f U (v,T,θ,d)+w2f S (v,T,θ,d,n,p,D,f,s,E,σ y ,ε)+w3f R (v,T,p,s)+w4f C (T, p, D, E);

[0062] Where w1, w2, w3, w4 are weight coefficients, satisfying And w i ≥0; The weight coefficient is determined by using the hierarchical analysis method or the fuzzy comprehensive evaluation method.

[0063] Furthermore, the method used to solve the multi-objective optimization model is a genetic algorithm.

[0064] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the above-mentioned method for determining parameters of a reinforcing bar winding machine.

[0065] A third aspect of the present invention provides a system for determining parameters of a reinforcing bar winding machine, which includes the above-mentioned computer-readable storage medium.

[0066] Compared with the prior art, the beneficial effects of the method, medium and system for determining parameters of a rib winding machine provided by the present invention are:

[0067] 1. A systematic parameter optimization model was established. This method establishes a mathematical model that reflects the relationship between the winding process parameters and the winding quality indicators, and adopts multi-objective optimization technology to obtain the optimal winding machine parameter combination that meets multiple quality requirements. This systematic parameter optimization method greatly improves the scientificity and controllability of the winding process, and utilizes

[0068] 2. It takes into account multiple rebar winding quality indicators. During the optimization process, this method takes into account multiple key quality indicators such as rebar winding uniformity, rebar winding strength, rebar winding surface finish, and rebar winding tightness. It can weigh the mutual influence of various indicators and seek the best balance point of various indicators. This multi-objective optimization method ensures the overall improvement of rebar winding quality.

[0069] In summary, the solution of the present invention solves the technical problem that the current process parameter determination of the rib winding machine mainly relies on experience accumulation and repeated experiments, and it is difficult to obtain the optimal process parameter combination. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 A flow chart of the method provided by the present invention. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0072] like Figure 1 FIG. 1 is a flow chart of a method for determining parameters of a reinforcing bar winding machine provided by the present invention. The method comprises the following steps:

[0073] S10, conduct multiple groups of rebar winding tests, collect rebar winding parameters and rebar winding quality under different process parameters and different rebar material parameters, the rebar winding parameters at least include rebar winding speed, rebar winding tension, rebar winding angle, rebar winding spacing, rebar winding number, rebar winding pressure, rebar winding quality at least includes rebar winding uniformity, rebar winding strength, rebar winding surface finish, rebar winding tightness, and rebar winding position accuracy; the rebar material parameters at least include rebar diameter, rebar strength grade, rebar surface state, rebar elastic modulus, rebar yield strength, and rebar elongation;

[0074] S20, establishing a group of equations for the quality of the winding process, including a winding uniformity equation, a winding strength equation, a winding surface finish equation, and a winding tightness equation;

[0075] S30, solving the process quality equation group to obtain multiple analytical solutions, namely, an analytical solution for rebar winding speed, an analytical solution for rebar winding tension, an analytical solution for rebar winding angle, an analytical solution for rebar winding spacing, an analytical solution for rebar winding number of turns, and an analytical solution for rebar winding pressure;

[0076] S40, establishing a multi-objective optimization model by taking the weighted sum of multiple analytical solutions as the objective function and taking the boundary conditions of the rib winding machine process and the material performance limitations as the constraint conditions;

[0077] S50, solving the multi-objective optimization model to obtain multiple groups of optimized process parameters;

[0078] S60, using the optimized process parameters to perform actual rib winding tests, and performing quality evaluation on the rib winding results, and recording the evaluation results of the rib winding quality indicators corresponding to each set of parameters;

[0079] S70. According to the evaluation results of the rib winding quality index, the best performing process parameters are selected as the optimal rib winding machine parameters and output.

[0080] The specific implementation methods of the above steps are described in detail below:

[0081] The specific implementation of step S10 is:

[0082] This step aims to obtain the relationship data between the performance of the rib winding machine and the process parameters and material parameters through systematic experiments, laying the foundation for the subsequent establishment of a mathematical model. The specific implementation process is as follows:

[0083] First, design the test plan. According to the working principle and process characteristics of the rebar winding machine, determine the process parameters and steel bar material parameters that need to be investigated. The process parameters include rebar winding speed v (m / min), rebar winding tension T (N), rebar winding angle θ (°), rebar winding spacing d (mm), number of rebar winding turns n and rebar winding pressure p (MPa). Rebar material parameters include rebar diameter D (mm), rebar strength grade f (MPa), rebar surface state s (dimensionless, the degree from smooth to rough can be represented by numbers 1-5), rebar elastic modulus E (GPa), rebar yield strength σ y (MPa) and steel bar elongation ε (%).

[0084] Secondly, determine the value range of the test parameters. For example, the rebar winding speed v can vary in the range of 5-50m / min, the rebar winding tension T can vary in the range of 100-1000N, the rebar winding angle θ can vary in the range of 30°-90°, the rebar winding spacing d can vary in the range of 5-50mm, the number of rebar winding turns n can vary in the range of 10-100, and the rebar winding pressure p can vary in the range of 0.1-1MPa. The rebar diameter D can be selected from common specifications such as 8, 10, 12, 14, 16, 18, 20, 22, 25, 28, 32mm, etc. The rebar strength grade f can be selected from 335, 400, 500, 600MPa and other levels. The rebar surface state s can be divided from 1 to 5. The rebar elastic modulus E is usually in the range of 200-210GPa, and the rebar yield strength σ y and elongation ε are determined according to the specific steel bar model.

[0085] Then, the orthogonal test design method is used to design multiple test plans. The orthogonal test method can obtain more comprehensive test data in a smaller number of tests and improve the test efficiency. According to the number and level of parameters, select an appropriate orthogonal table, such as L 16 (4 5 ) or L 32 (4 9 )wait.

[0086] Next, the rebar winding test is carried out. According to the designed test plan, the process parameters of the rebar winding machine are adjusted in turn, and the corresponding steel bar materials are selected for rebar winding. During the test, high-precision sensors and measuring equipment are used to collect rebar winding parameters and rebar winding quality data. The collection of rebar winding parameters can be achieved in the following ways: the rebar winding speed v is measured using a photoelectric encoder; the rebar winding tension T is measured using a tension sensor; the rebar winding angle θ is measured using an angle sensor or a visual system; the rebar winding spacing d is measured using a laser rangefinder or a visual system; the number of rebar winding turns n is obtained through a counter or a visual system; the rebar winding pressure p is measured using a pressure sensor.

[0087] The measurement method of rebar winding quality indicators is as follows: the rebar winding uniformity U can be characterized by measuring the standard deviation of the rebar winding spacing at multiple points. The smaller the standard deviation, the better the uniformity; the rebar winding strength S can be measured by a tensile test; the rebar winding surface finish R can be measured using a surface roughness meter; the rebar winding tightness C can be characterized by measuring the volume change rate of the steel bar after winding; the rebar winding position accuracy A can be determined by measuring the deviation between the actual rebar winding position and the theoretical position.

[0088] Finally, the collected data is preprocessed and statistically analyzed. Data cleaning techniques are used to remove outliers and noise, and interpolation methods are used to fill in missing data. The mean and standard deviation of each set of test data are calculated, and a correlation scatter plot between the parameters is drawn to preliminarily analyze the relationship between the parameters. The processed data is organized into a standard format to prepare for the subsequent establishment of a mathematical model.

[0089] The specific implementation of step S20 is:

[0090] This step aims to establish a mathematical model that describes the relationship between the quality of the rib winding process and various influencing factors, providing a theoretical basis for subsequent optimization. The specific implementation process is as follows:

[0091] Firstly, based on the test data and process mechanism, the influence of various process parameters and material parameters on the quality index of winding is analyzed. The equations of winding uniformity, winding strength, winding surface finish and winding tightness are established by using the multivariate regression analysis method.

[0092] The process of establishing the winding uniformity equation is as follows: Assuming that the winding uniformity U is mainly affected by the winding speed v, the winding tension T, the winding angle θ and the winding spacing d, it can be expressed as:

[0093] U=f U (v,T,θ,d)

[0094] Using the polynomial regression model and taking into account the second-order interaction terms, we can obtain:

[0095] U=a0+a1v+a2T+a3θ+a4d+a5v 2 +a6T 2 +a7θ 2 +a8d 2 +a9vT+a 10 vθ+a 11 vd+a 12 Tθ+a 13 Td+a 14 θd

[0096] Among them, a0, a1, …, a 14is the unknown coefficient. The test data is fitted using the least squares method to obtain the estimated values ​​of each coefficient. In order to avoid overfitting, stepwise regression or regularization methods (such as Lasso regression) can be used to screen important variables.

[0097] The process of establishing the reinforcement strength equation is similar. Considering that the reinforcement strength S is also closely related to the reinforcement material parameters, it can be expressed as:

[0098] S=f S (v,T,θ,d,n,p,D,f,s,E,σ y ,ε)

[0099] Parameter meaning:

[0100] S: Reinforcement strength, in MPa, indicating the tensile strength of the steel bar after rebar winding;

[0101] v: winding speed, in m / min, indicating the linear speed of the winding machine;

[0102] T: Rebar winding tension, in N, represents the tensile force applied to the steel bar during the winding process;

[0103] θ: Reinforcement angle, in degrees (°), indicating the angle between the steel bar and the axis of the reinforcement;

[0104] d: Reinforcement spacing, in mm, indicating the distance between two adjacent turns of reinforcement;

[0105] n: number of turns of reinforcement, no unit, represents the total number of turns of reinforcement;

[0106] p: Rebar winding pressure, in MPa, represents the pressure applied to the steel bar surface during the rebar winding process;

[0107] D: Steel bar diameter, in mm, indicating the diameter of the steel bar to be wound;

[0108] f: steel bar strength grade, in MPa, indicating the nominal yield strength of the steel bar;

[0109] s: steel bar surface condition, no unit, using integers from 1 to 5 to represent the degree from smooth to rough;

[0110] E: elastic modulus of steel bar, in GPa, indicating the stress-strain relationship of steel bar in the elastic stage;

[0111] σ y : Steel bar yield strength, in MPa, represents the stress at which the steel bar begins to yield;

[0112] ε: Steel bar elongation, expressed in %, which indicates the maximum plastic deformation of the steel bar when it breaks.

[0113] The polynomial regression model is also used. Considering that there are many variables, principal component analysis (PCA) can be performed to reduce the dimension, and then the regression equation can be established:

[0114]

[0115] Among them, PC1, PC2, …, PC m is the principal component, m is the number of principal components selected, and b0, b1,… are the coefficients to be determined.

[0116] Principal component analysis (PCA) is a dimensionality reduction technique used to reduce the number of variables while retaining most of the information. In this case, the principal components may include:

[0117] PC1=a 11 D+a 12 f+a 13 E+a 14 σ y +a 15 ε;

[0118] PC2=a 21 v+a 22 T+a 23 θ+a 24 d+a 25 n;

[0119] PC3=a 31 s+a 32 p; ...

[0120] Among them, a ij It is the coefficient obtained by principal component analysis, which indicates the weight of the original variable in the principal component. The specific coefficient value needs to be determined by performing PCA analysis on the actual data.

[0121] The calculation steps of PCA are well-known techniques and are briefly described as follows:

[0122] Standardize the data; calculate the covariance matrix; calculate the eigenvalues ​​and eigenvectors of the covariance matrix; select the eigenvectors corresponding to the largest k eigenvalues; project the data onto these k eigenvectors.

[0123] The surface finish equation of the rib can be expressed as:

[0124] R = f R (v,T,p,s)

[0125] Considering that surface finish may have a nonlinear relationship with these parameters, a generalized additive model (GAM) can be used:

[0126] R=c0+f1(v)+f2(T)+f3(p)+f4(s)

[0127] Among them, R represents the surface finish, usually expressed as Ra value (in μm); c0 is a constant term. f1(v), f2(T), f3(p), f4(s) are smooth functions, which respectively describe the nonlinear effects of winding speed, winding tension, winding pressure and steel bar surface state on surface finish.

[0128] The definition of these smooth functions adopts spline function, which is a well-known technology. Taking f1(v) as an example, its definition is:

[0129]

[0130] Among them, B i (v) is the basis function (usually B-spline or natural spline), β i are the coefficients to be determined, and k is the number of basis functions.

[0131] Fitting method:

[0132] 1. Select knots: Select an appropriate number of knots within the variable range.

[0133] 2. Build Basis Function: Build B-spline or natural spline basis functions using the selected nodes.

[0134] 3. Penalized fitting: Use penalty terms to control the smoothness of the function and avoid overfitting. Optimize the objective function:

[0135]

[0136] Among them, y j is the observed value, x ij is the value of the i-th variable for the j-th observation and λ is the smoothing parameter.

[0137] 4. Cross-validation: Use generalized cross-validation (GCV) or k-fold cross-validation to select the best smoothing parameter λ.

[0138] 5. Parameter estimation: Use penalized least squares to estimate the coefficient β i .

[0139] When fitting in the specific implementation process, the gam() function in the mgcv package of R language or the GAM class in the pyGAM library of Python can be used to implement it. The fitting process of f2(T), f3(p), and f4(s) is the same as that of f1(v).

[0140] The winding tightness equation can be expressed as:

[0141] C = f C (T,p,D,E)

[0142] Considering that there may be an exponential relationship between tightness, tension and pressure, a nonlinear regression model is used:

[0143]

[0144] Among them, d0, d1, …, d6 are unknown coefficients. Nonlinear least squares method is used for parameter estimation.

[0145] When establishing the equations, the following points should be noted:

[0146] 1. Data standardization: Before performing regression analysis, all variables are standardized to eliminate the dimensional effect.

[0147] 2. Multicollinearity diagnosis: Calculate the variance inflation factor (VIF). If VIF>10, it means there is serious multicollinearity and you need to consider deleting some variables or using methods such as ridge regression.

[0148] 3. Model diagnosis: Test the appropriateness of the model through residual analysis, QQ plot and other methods. If obvious heteroskedasticity or non-normality is found, consider using generalized linear models or nonparametric regression methods.

[0149] 4. Model validation: Use cross-validation to evaluate the predictive performance of the model and calculate the root mean square error (RMSE) and the coefficient of determination R 2 And other indicators.

[0150] 5. Model interpretation: Analyze the importance of each variable, which can be evaluated by standardized regression coefficients or relative importance indicators (such as the Lindeman-Merenda-Gold method).

[0151] Finally, the four established equations are combined into the winding process quality equation group:

[0152] U=f U (v,T,θ,d)

[0153] S=f S (v,T,θ,d,n,p,D,f,s,E,σ y ,ε)

[0154] R = f R (v,T,p,s)

[0155] C = f C (T,p,D,E)

[0156] This set of equations describes the relationship between the winding quality and various process parameters and material parameters, laying the foundation for subsequent parameter optimization.

[0157] The specific implementation of step S30 is:

[0158] This step aims to solve the set of equations for the quality of the rib winding process established in step S20 to obtain analytical solutions for each process parameter. Since the set of equations involves multiple variables and complex nonlinear relationships, it is difficult to solve directly. Therefore, numerical methods and approximate techniques are used to obtain analytical solutions. The specific implementation process is as follows:

[0159] First, simplify and reduce the order of the equations. Through sensitivity analysis, determine the parameters that have the most significant impact on each quality indicator, and ignore the parameters with less impact to reduce the complexity of the equation. For example, for the winding uniformity equation, it may be found that the winding speed v and the winding tension T have the most significant impact, so the equation can be simplified to:

[0160] U≈f U (v,T)

[0161] Secondly, the simplified equation is linearized using the Taylor expansion method. Taking the uniformity equation of the reinforcement winding as an example, a first-order Taylor expansion is performed near a certain working point (v0, T0):

[0162]

[0163] Then, the optimality condition is used to find the analytical solution of each parameter. For the uniformity of the winding, the optimal condition is and Solving these two equations, we can get the analytical solutions of the winding speed and winding tension:

[0164]

[0165]

[0166] Similarly, by processing the rib strength equation, rib surface finish equation and rib tightness equation, analytical solutions for other process parameters can be obtained.

[0167] For the reinforcement strength equation, assuming that it is mainly affected by the reinforcement angle θ and the number of reinforcement turns n after simplification, we can get:

[0168]

[0169]

[0170] For the surface finish equation of the rib winding, assuming that it is mainly affected by the rib winding pressure p, we can get:

[0171]

[0172] For the winding density equation, assuming that it is mainly affected by the winding spacing d, we can get:

[0173]

[0174] During the solution process, the following points should be noted:

[0175] 1. Selection of working point: The selection of working point (v0, T0, θ0, n0, p0, d0) has an important influence on the accuracy of analytical solution. You can choose the best performance parameter combination in historical data as the initial working point, or find the optimal working point through global optimization algorithm (such as particle swarm algorithm).

[0176] 2. Calculation of partial derivatives: For complex nonlinear functions, it may be difficult to directly obtain the partial derivatives in analytical form. In this case, numerical differentiation methods, such as the central difference method, can be used to approximate the partial derivatives. For example:

[0177]

[0178] Among them, h is a small step size, usually 10 -6 to 10 -8 The value between .

[0179] 3. Iterative solution: Since Taylor expansion is a local approximation near the working point, multiple iterative solutions may be required to obtain more accurate results. Use the Newton-Raphson method for iteration:

[0180] x k+1 =x k -[J(x k )] -1 F(x k )

[0181] Where x is the parameter vector, F(x) is the system of equations, and J(x) is the Jacobian matrix. The iteration termination condition can be set to ‖x k+1 -x k ‖<ε, ε is the preset convergence threshold, such as 10 -6 .

[0182] 4. Stability analysis of the solution: Perform stability analysis on the obtained analytical solution and calculate the eigenvalues ​​of the Jacobian matrix. If the real parts of all eigenvalues ​​are negative, the solution is stable.

[0183] 5. Physical meaning test of the solution: Check whether the analytical solution is physically reasonable. For example, the winding speed v should not be negative, and the winding angle θ should be between 0° and 180°. If an unreasonable solution appears, the model or constraints need to be readjusted.

[0184] 6. Error analysis: Calculate the error between the analytical solution and the numerical solution to evaluate the accuracy of the analytical solution. Relative error can be used:

[0185]

[0186] If the relative error exceeds a preset threshold (such as 5%), it is necessary to consider using a higher-order Taylor expansion or other approximation methods.

[0187] Finally, the analytical solutions of the parameters are organized into standard forms:

[0188] Analytical solution of winding speed: v * =f v (U,T,θ,d,n,p,D,f,s,E,σ y ,ε)

[0189] Analytical solution of the tension force of the reinforcement: T * =f T (U,v,θ,d,v,p,D,f,s,E,σ y ,ε)

[0190] Analytical solution of the reinforcing bar angle: θ * =f θ (S,v,T,d,n,p,D,f,s,E,σ y ,ε)

[0191] Analytical solution for the spacing between reinforcing bars: d * =f d (C,v,T,θ,n,p,D,f,s,E,σ y ,ε)

[0192] Analytical solution for the number of windings: n * =f n (S,v,T,θ,d,p,D,f,s,E,σ y ,ε)

[0193] Analytical solution of the reinforcement pressure: p * =f p (R,v,T,θ,d,n,D,f,s,E,σ y ,ε)

[0194] Among them, f v ,f T ,f θ ,f d ,f n ,f p are analytical functions of each parameter, which include the influence of other parameters and quality indicators. Please note that these formulas are approximate analytical solutions based on simplified models and local linearization.

[0195] 1.f v (Analytical function of winding speed):

[0196]

[0197] 2.f T (Analytical function of tendon tension):

[0198]

[0199] 3.f θ (Analytical function of winding angle):

[0200]

[0201] 4.f d (Analytical function of winding spacing):

[0202]

[0203] 5.f n (Analytical function of the number of winding turns):

[0204]

[0205] 6.f p (Analytical function of reinforcement pressure):

[0206]

[0207] In these formulas:

[0208] v0,T0,θ0,d0,n0,px are the parameter values ​​of the initial working point.

[0209] f U ,f S ,f R ,f C are functions describing uniformity, strength, surface finish and compactness respectively.

[0210] Represents the partial derivative of the function f with respect to the variable x.

[0211] U, S, R, and C represent the quality indicators of uniformity, strength, surface finish, and compactness, respectively.

[0212] D,f,s,E,σ y , ε are the steel bar material parameters, which represent the steel bar diameter, strength grade, surface condition, elastic modulus, yield strength and elongation respectively.

[0213] These analytical solutions provide quantitative relationships between process parameters and quality indicators, laying the foundation for the subsequent optimization process.

[0214] The specific implementation of step S40 is:

[0215] This step aims to establish a multi-objective optimization model based on the analytical solution obtained above to achieve the overall optimization of the winding process parameters. The specific implementation process is as follows:

[0216] First, construct the objective function. Considering multiple quality indicators of the rib winding process, the weighted sum method is used to combine multiple objectives into a comprehensive objective function:

[0217] F=w1f U (v,T,θ,d)+w2f S (v,T,θ,d,n,p,D,f,s,E,σ y ,ε)+w3f R (v,T,p,s)+w4f C (T,p,D,E)

[0218] Among them, w1, w2, w3, w4 are the weight coefficients of each quality index, satisfying And w i ≥ 0. The weight coefficient can be determined by using the analytic hierarchy process (AHP) or the fuzzy comprehensive evaluation method, combined with expert experience and actual production needs.

[0219] Secondly, determine the constraints. Constraints include boundary conditions of process parameters and material performance limitations:

[0220] 1. Reinforcement winding speed constraint: v min ≤v≤v max ;

[0221] 2. Tension constraint of reinforcement: T min ≤T≤T max ;

[0222] 3. Reinforcement angle constraint: θ min ≤θ≤θ max ;

[0223] 4. Reinforcement spacing constraint: d min ≤d≤d max ;

[0224] 5. Constraint on the number of winding circles: n min ≤n≤n max ;

[0225] 6. Reinforcement pressure constraint: p min ≤p≤p max ;

[0226] 7. Material strength constraint: S ≥ S required ;

[0227] 8. Surface finish constraint: R≤R max ;

[0228] 9. Closeness constraint: C ≥ C min ;

[0229] 10. Uniformity constraint: U≤U max .

[0230] The subscripts min and max represent the minimum and maximum values ​​of the parameters, respectively. These boundary values ​​can be determined according to the performance indicators of the winding machine and the actual process requirements. required is the required minimum strength, R max is the maximum allowable surface roughness, C min is the required minimum tightness, U max is the maximum allowable unevenness.

[0231] Then, consider the correlation constraints between parameters. For example, there may be a certain relationship between the winding speed and the winding tension:

[0232] g(v,T)≤0

[0233] Among them, g(v,T) is a function that describes the relationship between velocity and tension, which can be obtained by fitting experimental data.

[0234] Next, a multi-objective optimization model is established:

[0235] min F=w1f U (v,T,θ,d)+w2f S (v,T,θ,d,n,p,D,f,s,E,σ y ,ε)+w3f R (v,T,p,s)+w4f C (T,p,D,E)

[0236] Constraints include:

[0237] v min ≤v≤v max ; T min ≤T≤T max θ min ≤θ≤θ max ;d min ≤d≤d max ;n min ≤n≤n max ;p min ≤p≤p max ; S≥S required ; R≤R max ; C≥C min ; U≤U max ; g(v,T)≤0.

[0238] When establishing a multi-objective optimization model, the following points should be noted:

[0239] 1. Normalization of objective function: Since different quality indicators may have different dimensions and magnitudes, each objective function needs to be normalized to ensure that they are comparable during the optimization process. The maximum and minimum value normalization method can be used:

[0240]

[0241] Among them, f i ′ is the normalized objective function, f i,min and f i,max The objective function f i The minimum and maximum values ​​of .

[0242] 2. Sensitivity analysis of weight coefficients: Conduct sensitivity analysis on weight coefficients to evaluate the impact of weight changes on optimization results. Monte Carlo simulation method can be used to randomly generate multiple sets of weight coefficients and analyze the distribution of optimization results.

[0243] 3. Relaxation of constraints: For some constraints that are difficult to strictly meet, you can consider introducing relaxation variables to convert hard constraints into soft constraints. For example, rewrite the strength constraint as:

[0244] S+ξ≥S required ;ξ≥0;

[0245] Among them, ξ is a slack variable, and a penalty term Mξ is added to the objective function, where M is a large positive number.

[0246] 4. Robustness considerations: Considering the possible parameter fluctuations in actual production, a robust optimization method can be introduced. For example, a scenario-based robust optimization method is used to consider multiple possible scenarios:

[0247]

[0248] Among them, K is the number of scenarios considered, F i is the objective function in the i-th scenario.

[0249] 5. Multi-objective trade-off analysis: Use the concept of Pareto optimality, draw the Pareto frontier, and analyze the trade-offs between different objectives. Multi-objective optimization algorithms such as NSGA-II (Non-dominated Sorting Genetic Algorithm II) can be used to obtain the Pareto optimal solution set.

[0250] 6. Dynamic update of the model: Considering that the rib winding process may change over time, an online learning mechanism can be introduced to regularly update the model parameters. For example, recursive least squares (RLS) or Kalman filter can be used to update the model coefficients.

[0251] Finally, the established multi-objective optimization model is transformed into a standard mathematical programming form to prepare for the next step of solving. The model can be expressed as:

[0252]

[0253] stg i (x)≤0,i=1,2,…,m

[0254] h j (x)=0,j=1,2,…,n

[0255] x l ≤x≤x u

[0256] Among them, x is the decision variable vector, including all process parameters; F(x) is the objective function; g i (x) and h j (x) are inequality and equality constraints respectively; x l and x u are the lower and upper bounds of the decision variables, respectively.

[0257] This multi-objective optimization model comprehensively considers all aspects of the rib winding process and lays the foundation for the next step of solving the optimal process parameters.

[0258] The specific implementation of step S50 is:

[0259] This step aims to solve the multi-objective optimization model established in step S40 and obtain multiple sets of optimized process parameters. Due to the complexity and nonlinear characteristics of the model, an intelligent optimization algorithm is used to solve it. The specific implementation process is as follows:

[0260] First, select an appropriate optimization algorithm. Considering the characteristics of the problem, you can choose the following algorithms:

[0261] 1. Genetic Algorithm (GA): Suitable for dealing with complex nonlinear optimization problems and has good global search capabilities.

[0262] 2. Particle Swarm Optimization (PSO): It has fast convergence speed and is suitable for dealing with continuous variable optimization problems.

[0263] 3. Simulated annealing algorithm (SA): It can escape from the local optimal solution and is suitable for dealing with problems with multiple local optimal points.

[0264] 4. Differential Evolution (DE): It is insensitive to parameters and suitable for high-dimensional optimization problems.

[0265] 5. Ant Colony Algorithm (ACO): Suitable for dealing with combinatorial optimization problems and can be used for discrete process parameter optimization.

[0266] Here we take the genetic algorithm as an example to explain the solution process in detail:

[0267] 1. Coding: Encode the decision variables (process parameters) into chromosomes. Use real number coding, and each gene represents a process parameter. The chromosome structure is:

[0268] [v,T,θ,d,n,p]

[0269] 2. Initialize the population: Generate N individuals to form the initial population. The gene value of each individual is randomly generated within its corresponding constraint range. The population size N can be set to 50-200, and the specific value needs to be weighed according to the problem scale and computing resources.

[0270] 3. Fitness evaluation: Calculate the fitness value of each individual. The fitness function is:

[0271]

[0272] Among them, F(x) is the objective function in the multi-objective optimization model.

[0273] 4. Selection operation: Use the roulette wheel selection method to select outstanding individuals to enter the next generation. The selection probability is proportional to the fitness of the individual.

[0274] 5. Crossover operation: Select two parent individuals and perform crossover operation to generate offspring. Use the simulated binary crossover (SBX) method:

[0275] c1=0.5[(1+β)p1+(1-β)p2]

[0276] c2=0.5[(1-β)p1+(1+β)p2]

[0277] Among them, c1 and c2 are offspring, p1 and p2 are parents, and β is the crossover factor, which is generated by the following distribution:

[0278]

[0279] Among them, u is a random number between [0,1], η c is the distribution index, usually ranging from 20 to 100.

[0280] 6. Mutation operation: Randomly perturb certain genes of individuals. Use polynomial mutation method:

[0281] c=p+δ(p u -p l )

[0282] Among them, c is the gene value after mutation, p is the gene value before mutation, and p uand p l are the upper and lower bounds of the gene, respectively, and δ is the mutation factor, which is generated by the following distribution:

[0283]

[0284] Among them, u is a random number between [0,1], η m is the distribution index, usually ranging from 20 to 100.

[0285] 7. Constraint processing: For individuals that do not meet the constraint conditions, the penalty function method is used for processing. The modified fitness function is:

[0286]

[0287] Among them, r i and r j is the penalty coefficient, which can be increased as the number of iterations increases to enhance the impact of the constraint.

[0288] 8. Elite retention: directly copy the best individuals in the current population to the next generation to ensure that excellent genes are not lost.

[0289] 9. Termination condition: Set the maximum number of iterations (such as 1000) or the optimal solution for multiple consecutive generations without significant improvement as the termination condition.

[0290] 10. Result output: Output the non-dominated solution set in the final population, that is, multiple sets of optimized process parameters.

[0291] During implementation, the following points should be noted:

[0292] 1. Parameter tuning: The performance of genetic algorithms is affected by multiple parameters, such as population size, crossover probability, mutation probability, etc. These parameters can be optimized using orthogonal test method or response surface method.

[0293] 2. Multiple runs: Due to the randomness of genetic algorithms, it is recommended to run the algorithm multiple times (such as 30 times) to obtain the best result or perform statistical analysis.

[0294] 3. Local search: On the basis of genetic algorithm, local search strategies such as pattern search or gradient descent can be introduced to improve the accuracy of the solution. This method of combining global search and local search is called hybrid genetic algorithm (HybridGA).

[0295] 4. Parallel computing: Use parallel computing technology to accelerate the optimization process. A master-slave parallel model can be used to assign fitness evaluation tasks to multiple slave processors.

[0296] 5. Dynamic adjustment strategy: Dynamically adjust algorithm parameters during the optimization process. For example, as the number of iterations increases, the probability of mutation can be gradually reduced to enhance local search capabilities.

[0297] 6. Diversity maintenance: In order to prevent the population from converging prematurely, crowding control or microhabitat technology can be used to maintain the diversity of the population.

[0298] 7. Convergence analysis: Analyze the convergence characteristics of the algorithm by plotting the change curve of the objective function value with the number of iterations. If it is found that the convergence is too fast or too slow, the algorithm parameters need to be adjusted accordingly.

[0299] 8. Sensitivity analysis: Conduct sensitivity analysis on the optimization results to evaluate the influence of each process parameter on the objective function. This analysis can be performed using analysis of variance (ANOVA) or partial differential analysis.

[0300] Finally, post-process and analyze the optimization results:

[0301] 1. Solution screening: If the obtained non-dominated solution set is large, cluster analysis methods (such as k-means algorithm) can be used to classify the solutions and select representative solutions.

[0302] 2. Decision support: Use visualization techniques (such as scatter plot matrix and parallel coordinates plot) to display multi-objective optimization results and help decision makers choose the most suitable combination of process parameters.

[0303] 3. Robustness analysis: Conduct robustness analysis on the optimization results to evaluate the impact of small fluctuations in process parameters on quality indicators. Monte Carlo simulation methods can be used to generate random perturbations of parameters and analyze the range of variation of quality indicators.

[0304] 4. Interactive optimization: Develop an interactive optimization interface that allows users to adjust the weights or constraints of the objective function according to actual needs and observe changes in optimization results in real time.

[0305] Through the above steps, multiple sets of optimized process parameters can be obtained, which can theoretically achieve the comprehensive optimization of the winding quality. This provides a theoretical basis and reference for the next step of practical verification.

[0306] The specific implementation of step S60 is:

[0307] This step aims to verify the effectiveness of the optimized process parameters through actual rib winding tests and to quantitatively evaluate the rib winding results. The specific implementation process is as follows:

[0308] 1. Test preparation:

[0309] a) Designing an experimental scheme according to the multiple groups of optimized process parameters obtained in step S50. In order to improve the experimental efficiency, a partial factorial experimental design method, such as an orthogonal experimental method or a central composite experimental design, may be used.

[0310] b) Prepare test equipment, including rib winding machines, various sensors (such as force sensors, displacement sensors, angle sensors, etc.), data acquisition systems, etc.

[0311] c) Prepare test materials, including steel bars of different specifications and strength grades.

[0312] d) Calibrate all measuring equipment to ensure measurement accuracy.

[0313] 2. Experimental implementation:

[0314] a) According to the test plan, set the process parameters of the rib winding machine (rib winding speed, rib winding tension, rib winding angle, rib winding spacing, rib winding number, rib winding pressure) in sequence.

[0315] b) Select steel bars of appropriate specifications and carry out the reinforcement winding operation.

[0316] c) During the winding process, sensors are used to collect various parameter data in real time, including actual winding speed, tension, angle, etc.

[0317] d) For each set of parameters, the test was repeated 3-5 times to reduce the impact of random errors.

[0318] 3. Quality Assessment:

[0319] The following quality indicators are measured and evaluated for each rib wound sample:

[0320] a) Uniformity of winding (U):

[0321] Use a high-precision coordinate measuring machine or laser scanner to measure the spacing between the ribs.

[0322] Calculate the standard deviation of the spacing; the smaller the standard deviation, the better the uniformity.

[0323] Define uniformity metrics: where σ d is the standard deviation of the interval, and d is the average interval.

[0324] b) Reinforcement strength (S):

[0325] The tensile test was carried out using a universal testing machine.

[0326] The yield strength and tensile strength were recorded.

[0327] Calculate the strength retention rate: where f y is the yield strength after reinforcement, fy0 is the yield strength before reinforcement.

[0328] c) Surface finish of the ribs (R):

[0329] The roughness of the rib surface was measured using a surface roughness tester.

[0330] Record the arithmetic mean deviation value.

[0331] Define the finish index: Among them, Ra max is the maximum allowed roughness value.

[0332] d) Reinforcement tightness (C):

[0333] Measure the volume change of the steel bars before and after winding.

[0334] Calculate the compactness index: Where V0 is the volume before winding, and V is the volume after winding.

[0335] e) Reinforcement position accuracy (A):

[0336] Use a laser tracker or vision system to measure the deviation between the actual rib winding position and the theoretical position.

[0337] Calculate the position accuracy index: where Δx i ,Δy i ,Δz i is the deviation of the ith measuring point in three directions, n is the number of measuring points, and L is the total length of the rebar.

[0338] 4. Data processing and analysis:

[0339] a) Perform statistical analysis on the repeated test results of each group of parameters and calculate the mean value and standard deviation of each quality indicator.

[0340] b) Perform analysis of variance (ANOVA) to evaluate the significance of the effects of different process parameters on various quality indicators.

[0341] c) Establish a regression model between process parameters and quality indicators. You can consider using methods such as response surface methodology (RSM) or artificial neural network (ANN).

[0342] d) To calculate the comprehensive score of each quality indicator, the weighted sum method can be used:

[0343] Score=w1U+w2S+w3R+w4C+w5A

[0344] Among them, w1, w2, w3, w4, and w5 are the weights of each indicator, which can be determined by the analytic hierarchy process (AHP).

[0345] 5. Result evaluation and recording:

[0346] a) Comprehensively evaluate the test results of each set of process parameters and record the specific values ​​and comprehensive scores of each quality indicator.

[0347] b) Draw a relationship diagram between process parameters and quality indicators, such as a scatter plot, contour plot, etc., to intuitively show the impact of parameters on quality.

[0348] c) Write a detailed test report, including test methods, data analysis process, results discussion, etc.

[0349] 6. Abnormal situation handling:

[0350] a) If any abnormal situation occurs during the test, such as steel bar breakage, rebar winding machine failure, etc., it is necessary to record the abnormal situation and possible causes in detail.

[0351] b) For abnormal data, use statistical methods (such as Grubbs criterion) to test outliers and decide whether to exclude them.

[0352] c) If the test results of a certain set of parameters deviate significantly from the theoretical predictions, repeated tests are required to verify and analyze the possible causes.

[0353] 7. Parameter optimization feedback:

[0354] a) Based on the test results, the multi-objective optimization model established in step S40 is corrected and calibrated.

[0355] b) If the test results show that some parameter combinations are not ideal, new parameter combinations can be predicted through interpolation or extrapolation methods and additional tests can be carried out.

[0356] 8. Process parameter sensitivity analysis:

[0357] a) Analyze the sensitivity of each process parameter to the quality of winding by controlling the variable method.

[0358] b) Calculate the sensitivity coefficient of each parameter: Where Q is the quality index, P i is the ith process parameter.

[0359] c) Draw a sensitivity radar chart to intuitively display the importance of each parameter.

[0360] 9. Process window analysis:

[0361] a) Based on the test results, determine the optimal range of each process parameter.

[0362] b) Draw a process window diagram to show the parameter combination area that meets quality requirements.

[0363] c) Analyze the width of the process window and evaluate the stability and controllability of the process.

[0364] 10. Reliability assessment:

[0365] a) Carry out repeated tests (e.g. 30 times) on the optimal process parameters to evaluate the stability of the winding quality.

[0366] b) Calculate the process capability index (Cpk) of each quality indicator and evaluate the reliability of the process.

[0367] c) Conduct life tests to evaluate the long-term performance of the rebar wound products.

[0368] Through the above steps, the actual effect of the optimized process parameters can be comprehensively evaluated, and the evaluation results of the winding quality indicators corresponding to each set of parameters can be obtained. These results will provide an important basis for the final determination of the optimal winding machine parameters.

[0369] The specific implementation of step S70 is:

[0370] This step aims to select the best performing process parameters as the optimal parameters of the rib winding machine according to the evaluation results of the rib winding quality index obtained in step S60, and output the final results. The specific implementation process is as follows:

[0371] 1. Data aggregation and standardization:

[0372] a) Collect the quality index evaluation results of all test groups in step S60, including rib winding uniformity (U), rib winding strength (S), rib winding surface finish (R), rib winding tightness (C) and rib winding position accuracy (A).

[0373] b) Standardize each indicator to eliminate the dimension effect. Use the maximum and minimum value standardization method:

[0374]

[0375] Among them, X is the original index value, X norm is the standardized value.

[0376] 2. Weight determination:

[0377] a) Use the analytic hierarchy process (AHP) to determine the weight of each quality indicator.

[0378] b) Construct a judgment matrix and invite experts in winding technology to compare and score the two products.

[0379] c) Calculate the eigenvector and obtain the weights w1, w2, w3, w4, w5 of each indicator.

[0380] d) Perform a consistency check to ensure that the consistency ratio CR of the judgment matrix is ​​less than 0.1.

[0381] 3. Comprehensive score calculation:

[0382] For each set of process parameters, calculate its comprehensive score:

[0383] Score = w1U norm +w2S norm +w3R norm +w4C norm +w5A norm

[0384] Among them, U norm ,S norm ,R norm ,C norm ,A norm are the standardized quality index values.

[0385] 4. Sorting and filtering:

[0386] a) Arrange all process parameter combinations in descending order according to the comprehensive scores.

[0387] b) Select the top five groups of process parameters with the highest comprehensive scores for further analysis.

[0388] 5. Stability analysis:

[0389] a) Carry out repeatability test on the selected 5 groups of process parameters (e.g. repeat 10 times for each group).

[0390] b) Calculate the coefficient of variation CV of the quality index of each group of parameters:

[0391]

[0392] Among them, σ is the standard deviation and μ is the mean.

[0393] c) Considering the score and stability comprehensively, a weighted method can be used:

[0394] FinalScore=Score×(1-αCV)

[0395] Among them, α is the trade-off coefficient, which can be 0.1-0.3.

[0396] 6. Optional, process window analysis:

[0397] a) Conduct process window analysis on the selected excellent process parameter combinations.

[0398] b) Using the two most important process parameters as coordinate axes, draw a contour map to show the trend of quality indicators as the parameters change.

[0399] c) Determine the parameter range that meets all quality requirements, i.e. the process window.

[0400] 7. Optional, sensitivity analysis:

[0401] a) Conduct local sensitivity analysis on the optimal process parameters.

[0402] b) Calculate the sensitivity coefficient of each parameter:

[0403]

[0404] Among them, P i is the ith process parameter.

[0405] c) Draw a sensitivity bar graph to visually display the importance of each parameter.

[0406] 8. Optional, economic analysis:

[0407] a) Calculate the production efficiency and energy consumption under different combinations of process parameters.

[0408] b) Estimate material costs, equipment depreciation, labor costs, etc.

[0409] c) Conduct a cost-benefit analysis and calculate the input-output ratio of each set of parameters.

[0410] 9. Final parameter determination:

[0411] a) Comprehensively consider the quality score, stability, process window, sensitivity and economy to select the optimal combination of process parameters.

[0412] b) If the performance of multiple groups of parameters is similar, a weighted decision can be made:

[0413] Decision=β1FinalScore+β2WindowSize+β3Robustness+β4Efficiency

[0414] Among them, β1, β2, β3, and β4 are the weights of each factor.

[0415] 10.Optional, result verification and confirmation:

[0416] a) Carry out large-scale trial production (e.g. 100 pieces) using the selected optimal process parameters.

[0417] b) Carry out comprehensive quality inspection, including destructive testing and non-destructive testing.

[0418] c) Calculate the process capability index Cpk of each quality indicator and ensure that Cpk>1.33.

[0419] d) Conduct long-term performance evaluation, such as fatigue test, corrosion resistance test, etc.

[0420] A second aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the above-mentioned method for determining parameters of a reinforcing bar winding machine.

[0421] A third aspect of the present invention provides a system for determining parameters of a reinforcing bar winding machine, which includes the above-mentioned computer-readable storage medium.

[0422] Specifically, the principle of the present invention is to establish a mathematical model reflecting the relationship between the winding process parameters and the winding quality index, and based on this, use multi-objective optimization technology to obtain the optimal parameter combination.

[0423] First, through the collection of a large amount of rebar winding test data, the rebar winding parameters and rebar winding quality indicators under different process parameters and steel bar material parameters are obtained. These data provide a basis for subsequent modeling and optimization.

[0424] Secondly, according to the test data, the quality equations of the winding process are established, including the winding uniformity equation, the winding strength equation, the winding surface finish equation and the winding tightness equation. These equations reflect the mathematical relationship between the winding process parameters, such as winding speed, winding tension, winding angle, etc., and the winding quality indicators. By establishing these equations, the mathematical model foundation for parameter optimization is laid.

[0425] Then, the quality equations of the winding process are solved to obtain the optimal analytical solutions of various winding process parameters. These analytical solutions provide key inputs for subsequent multi-objective optimization.

[0426] Next, a multi-objective optimization model is established. The weighted sum of the analytical solutions of the rebar winding process parameters obtained in step 3 is used as the objective function. At the same time, the process boundary conditions of the rebar winding machine and the performance restrictions of the steel bar material are used as constraints. Multiple sets of optimized rebar winding process parameters are obtained through the optimization algorithm. This multi-objective optimization method can find the optimal parameter combination for the comprehensive rebar winding quality index under the premise of meeting various process and material constraints.

[0427] Finally, the actual winding test was carried out using the optimized winding process parameters to measure and evaluate the winding quality indicators. Based on the test results, the parameter combination with the best comprehensive performance was selected as the final optimal winding machine parameters on the premise of meeting the quality index threshold requirements.

[0428] In general, the key to the method of the present invention is to systematically analyze and solve the complex relationship between the winding process parameters and the winding quality through mathematical modeling and multi-objective optimization, and obtain the optimal parameter combination that meets multiple quality index requirements.

[0429] In order to better understand and implement the present invention, a specific embodiment of the present invention is provided below: The following is a specific embodiment of the present invention, which describes in detail the application process of a method for determining parameters of a tendon winding machine:

[0430] In a prestressed concrete component production plant, in order to improve the quality and production efficiency of the rebar winding, it is decided to adopt the rebar winding machine parameter determination method of the present invention. The plant mainly produces prestressed concrete poles, and the steel bar specifications used are HPB300 grade Φ8mm threaded steel bars.

[0431] First, multiple sets of rebar winding tests were performed according to step S10. An orthogonal test scheme was designed, taking into account six process parameters (rebar winding speed, rebar winding tension, rebar winding angle, rebar winding spacing, number of rebar winding turns, and rebar winding pressure) and six steel bar material parameters (rebar diameter, rebar strength grade, rebar surface state, rebar elastic modulus, rebar yield strength, and rebar elongation).

[0432] In order to reduce the number of experiments, the L36 orthogonal table was used. This orthogonal table allows multiple factors to be considered simultaneously, each factor has 2 or 3 levels, and a total of 36 experiments can be arranged. Based on experience and equipment limitations, the value range of each parameter was determined:

[0433] 1. Reinforcement winding speed: 10-30m / min

[0434] 2. Reinforcement tension: 200-600N

[0435] 3. Reinforcement angle: 45°-75°

[0436] 4. Reinforcement spacing: 10-30mm

[0437] 5. Number of winding circles: 20-60

[0438] 6. Reinforcement pressure: 0.2-0.6MPa

[0439] The steel bar material parameters are determined based on the actual performance of HPB300 grade Φ8mm threaded steel bars.

[0440] The test was conducted using high-precision sensors and measuring equipment. The winding speed was measured by a photoelectric encoder, the winding tension was measured by a tension sensor, the winding angle and spacing were measured by a visual system, the number of windings was obtained by a counter, and the winding pressure was measured by a pressure sensor.

[0441] The following methods are used to measure the quality index of the reinforcement:

[0442] 1. Reinforcement uniformity: Use a laser rangefinder to measure the spacing between 20 points and calculate the standard deviation.

[0443] 2. Rebar winding strength: Cut standard specimens from the rebar after winding and carry out tensile test.

[0444] 3. Surface finish of the rib: Use a surface roughness meter to measure the Ra value.

[0445] 4. Rebar winding density: measure the volume change rate of the steel bars before and after winding.

[0446] 5. Reinforcement winding position accuracy: Use a three-dimensional coordinate measuring machine to measure the deviation between the actual reinforcement winding position and the theoretical position.

[0447] After completing the 36 experiments, the collected data were preprocessed and statistically analyzed. The Python pandas library was used to clean the data, remove obvious outliers, and use interpolation to fill in a small amount of missing data. Then, the seaborn library was used to draw a heat map of the correlation between the parameters and preliminarily analyze the relationship between the parameters.

[0448] Next, we enter step S20 and start to establish the quality equations of the rib winding process. We use the Python statsmodels library to perform multivariate regression analysis and establish four equations:

[0449] 1. Reinforcement uniformity equation:

[0450] It was found that the uniformity of winding is mainly affected by winding speed, winding tension and winding spacing. A second-order polynomial regression model was used, considering the main effects, square terms and interaction terms of these three variables.

[0451] 2. Reinforcement strength equation:

[0452] The strength of the ribs is related to all process parameters and material parameters. Due to the large number of variables, principal component analysis (PCA) was first used to reduce the dimension, and then a regression equation was established. This process was implemented using the PCA module of the sklearn library.

[0453] 3. Surface finish equation of rib winding:

[0454] It was found that the surface finish was mainly affected by the winding speed, winding pressure and the surface condition of the reinforcement. A generalized additive model (GAM) was used to capture the nonlinear relationship between these variables and the surface finish. The pygam library of Python was used to implement GAM.

[0455] 4. Reinforcement tightness equation:

[0456] The compactness is mainly affected by the winding tension, winding pressure, steel bar diameter and steel bar elastic modulus. A nonlinear regression model is used to estimate parameters using the curve_fit function in the scipy.optimize module.

[0457] When developing these equations, we took into account the problem of multicollinearity. We calculated the variance inflation factor (VIF), and processed the variables with a VIF greater than 10, such as merging correlated variables or using ridge regression.

[0458] To verify the appropriateness of the model, residual analysis was performed and QQ plots were drawn to test the normality of the residuals. A 5-fold cross validation was also used to evaluate the predictive performance of the model and the root mean square error (RMSE) and coefficient of determination R were calculated. 2 .

[0459] In step S30, the established equation group is solved to obtain analytical solutions of various process parameters. Since the equation group involves complex nonlinear relationships, numerical methods and approximation techniques are used.

[0460] Firstly, the equations were simplified and reduced in order. Through sensitivity analysis, the most significant parameters affecting each quality index were determined, while the parameters with less significant effects were ignored.

[0461] Then, the simplified equation is linearized by Taylor expansion method. The best parameter combination in historical data is selected as the initial working point, and the first-order Taylor expansion is performed near this point.

[0462] The Python SymPy library is used for symbolic calculations to obtain partial derivative expressions of each parameter. For complex nonlinear functions that are difficult to directly obtain analytical partial derivatives, numerical differentiation methods such as the central difference method are used.

[0463] Next, the optimality condition is used to solve the analytical solution of each parameter. The Newton-Raphson method is used for iterative solution, and the iteration termination condition is set to the difference between the results of two consecutive iterations is less than 10^-6.

[0464] After obtaining the analytical solution, the stability analysis of the solution is performed. The eigenvalues ​​of the Jacobian matrix are calculated to ensure that the real parts of all eigenvalues ​​are negative to ensure the stability of the solution. The physical meaning of the solution is also checked to ensure that all parameters are within a reasonable range.

[0465] In step S40, a multi-objective optimization model is established with the weighted sum of multiple analytical solutions as the objective function. According to actual production, the following constraints are set:

[0466] 1. The winding speed shall not exceed the maximum speed of the equipment 35m / min

[0467] 2. The tension of the steel bar should not exceed 80% of the yield strength of the steel bar.

[0468] 3. The winding angle is between 30° and 90°

[0469] 4. The spacing between the steel bars should not be less than 1.2 times the diameter of the steel bars.

[0470] 5. The number of winding circles meets the design requirements

[0471] 6. The winding pressure shall not exceed the maximum pressure of the equipment 0.8MPa

[0472] Use Python’s pyomo library to construct and solve this multi-objective optimization problem.

[0473] In step S50, the multi-objective optimization model is solved. A weighting method is used to convert multiple objective functions into a single objective function. The selection of weights is based on the importance of each quality indicator and is determined through discussions with production managers and quality control personnel. A genetic algorithm is used to solve this optimization problem because the genetic algorithm has a good effect on dealing with nonlinear, multi-constrained optimization problems.

[0474] Multiple sets of optimized process parameters were obtained, and five sets of parameters were selected for the next step of actual verification.

[0475] In step S60, the actual rib winding test was carried out using the five sets of optimized process parameters. Each set of parameters was tested three times to reduce the influence of random errors. A comprehensive quality evaluation of the rib winding results was carried out according to the previously established method, and the rib winding uniformity, rib winding strength, rib winding surface finish, rib winding tightness and rib winding position accuracy corresponding to each set of parameters were recorded.

[0476] Finally, in step S70, based on the evaluation results of the winding quality indicators, the best performing process parameters are selected as the optimal winding machine parameters. A comprehensive scoring method is adopted to assign different weights to each quality indicator and calculate the comprehensive score of each group of parameters. The determination of the weights takes into account product performance requirements and customer feedback, among which the winding strength and uniformity are assigned higher weights.

[0477] The parameter combination with the highest comprehensive score was determined as the optimal parameters of the rib winding machine, as follows:

[0478] 1. Reinforcement winding speed: 22m / min

[0479] 2. Reinforcement tension: 450N

[0480] 3. Reinforcement angle: 60°

[0481] 4. Reinforcement spacing: 15mm

[0482] 5. Number of winding circles: 40

[0483] 6. Reinforcement pressure: 0.4MPa

[0484] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for determining parameters of a tendon winding machine, characterized in that: The following steps are involved: S10, performing multiple groups of rebar winding tests, collecting rebar winding parameters and rebar winding quality under different process parameters and different rebar material parameters, wherein the rebar winding parameters at least include rebar winding speed, rebar winding tension, rebar winding angle, rebar winding spacing, rebar winding number of turns, and rebar winding pressure, and the rebar winding quality at least includes rebar winding uniformity, rebar winding strength, rebar winding surface finish, rebar winding tightness, and rebar winding position accuracy; the rebar material parameters at least include rebar diameter, rebar strength grade, rebar surface state, rebar elastic modulus, rebar yield strength, and rebar elongation; S20, establishing a group of equations for the quality of the winding process, including a winding uniformity equation, a winding strength equation, a winding surface finish equation, and a winding tightness equation; S30, solving the process quality equation group to obtain multiple analytical solutions, which are respectively an analytical solution for rebar winding speed, an analytical solution for rebar winding tension, an analytical solution for rebar winding angle, an analytical solution for rebar winding spacing, an analytical solution for rebar winding number of turns, and an analytical solution for rebar winding pressure; S40, establishing a multi-objective optimization model by taking the weighted sum of multiple analytical solutions as the objective function and taking the boundary conditions of the winding machine process and the material performance limitations as the constraints; S50, solving the multi-objective optimization model to obtain multiple groups of optimized process parameters; S60, using the optimized process parameters to perform actual rib winding tests, and performing quality evaluation on the rib winding results, and recording the evaluation results of the rib winding quality indicators corresponding to each set of parameters; S70. According to the evaluation result of the rib winding quality index, the best performing process parameters are selected as the optimal rib winding machine parameters and outputted.

2. A method for determining parameters of a rib winding machine according to claim 1, characterized in that: The uniformity equation of the reinforcement winding is specifically: U=f U (v,T,θ,d); The polynomial regression method is used, which is specifically expressed as: U=a0+a1v+a2T+a3θ+a4d+a5v 2 +a6T 2 +a7θ 2 +a8d 2 +a9vT+a 10 vθ+a 11 vd+a 12 Tθ+a 13 Td+a 14 θd; Where a0, a1, ..., a 14 is the unknown coefficient, v is the winding speed, T is the winding tension, θ is the winding angle, and d is the winding spacing.

3. A method for determining parameters of a rib winding machine according to claim 2, characterized in that: The strength equation of the reinforcing bar is specifically: S=f S (v,T,θ,d,n,p,D,f,s,E,p y ,e); Where S is the strength of the steel bar, v is the speed of the steel bar, n is the number of turns, p is the pressure of the steel bar, D is the diameter of the steel bar, f is the strength grade of the steel bar, s is the surface state of the steel bar, E is the elastic modulus of the steel bar, σ y is the yield strength of the steel bar, and ε is the elongation of the steel bar. The above formula adopts the polynomial regression method. Considering that there are many variables, the principal component analysis method is first used to reduce the dimension, and then the following regression equation is established: Where PC1, PC2, ..., PC m is the principal component, m is the number of principal components selected, and b0, b1, ... are the coefficients to be determined.

4. A method for determining parameters of a tendon winding machine according to claim 3, characterized in that: The surface finish equation of the rib winding is specifically: R=f R (v,T,p,s); Considering the nonlinear relationship between surface finish and the winding speed v, winding tension T, winding pressure p, and steel bar surface state s, a smooth function is used for fitting, as follows: R=c0+f1(v)+f2(T)+f3(p)+f4(s); Where f1(v), f2(T), f3(p), and f4(s) are smooth functions, which respectively describe the nonlinear effects of winding speed, winding tension, winding pressure, and steel bar surface state on surface finish, and are fitted using spline function.

5. A method for determining parameters of a tendon winding machine according to claim 4, characterized in that: The winding tightness equation is specifically: C=f C (T,p,D,E); Considering the exponential relationship between tightness, tension and pressure, a nonlinear regression model is used, as follows: Wherein, d0, d1, ..., d6 are unknown coefficients.

6. A method for determining parameters of a tendon winding machine according to claim 5, characterized in that: The analytical solution of the reinforcing bar winding speed is specifically: In the formula, v * is the optimal winding speed, v0 is the initial winding speed, and T0 is the initial winding tension; The analytical solution of the reinforcing bar tension is specifically: Where, T * is the optimal tendon tension; The analytical solution of the reinforcing bar angle is specifically: In the formula, θ * is the optimal reinforcement winding angle, θ0 is the initial reinforcement winding angle, and n0 is the initial number of reinforcement winding turns; The analytical solution for the spacing between reinforcing bars is specifically: Where d* is the optimal spacing between reinforcement bars, and d0 is the initial spacing between reinforcement bars; The analytical solution for the number of reinforcing bar turns is specifically: Where n * is the optimal number of reinforcing bar turns, n0 is the initial number of reinforcing bar turns; The analytical solution of the reinforcing bar pressure is specifically: In the formula, p * is the optimal reinforcement pressure, and p0 is the initial reinforcement pressure.

7. A method for determining parameters of a tendon winding machine according to claim 6, characterized in that: The objective function of the multi-objective optimization model is specifically: F=w1f U (v, T, θ, d)+w2f S (v, T, θ, d, n, p, D, f, s, E, σ y ,ε)+w3f R (v, T, p, s)+w4f C (T, p, D, E); In the formula, w1, w2, w3, w4 are weight coefficients, satisfying And w i ≥0; The weight coefficient is determined by using the hierarchical analysis method or the fuzzy comprehensive evaluation method.

8. A method for determining parameters of a tendon winding machine according to claim 7, characterized in that: The method used to solve the multi-objective optimization model is a genetic algorithm.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program instructions, and when the program instructions are executed, they are used to execute the method for determining parameters of a reinforcing bar winding machine according to any one of claims 1 to 8.

10. A parameter determination system for a tendon winding machine, characterized in that: A computer-readable storage medium comprising the computer-readable storage medium of claim 9.

Images