Method, device and system for optimizing cable force of extradosed cable-stayed bridge and storage medium

Abstract

The invention discloses an extradosed cable-stayed bridge cable force optimization method, device and system and a storage medium. The method comprises the steps that S1, cable force and bending strain energy training samples of an extradosed cable-stayed bridge are obtained; s2, optimizing SVM parameters under the cable force and bending strain energy training samples of the extradosed cable-stayed bridge by adopting an IGWO algorithm; and S3, taking the cable force value after secondary tensioning as a to-be-optimized parameter, carrying out iterative optimization by using an IGWO algorithm, taking the cable force combination output after evolution of each generation as an input vector of an SVM for nonlinear prediction, outputting a population fitness value under the current cable force, and repeatedly reducing the time cost when the optimization model carries out fitness evaluation so as to improve the fitness evaluation accuracy. And IGWO-based combination optimization is realized. By the adoption of the technical scheme, the calculation efficiency and calculation precision of the finished cable force optimization of the extradosed cable-stayed bridge are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of cable force optimization, and in particular relates to a cable force optimization method and device, a system, and a storage medium for a low-tower cable-stayed bridge. Background Art

[0002] Inadequate optimization of cable force of cable-stayed bridges and various errors during construction control will have a significant impact on the reasonable state of bridge construction. Therefore, it is of outstanding practical significance to study the optimization method of cable force of cable-stayed bridges and conduct precise construction monitoring. At present, the optimization method of cable force of cable-stayed bridges based on intelligent algorithms has greatly improved the calculation efficiency, but the research objects are basically conventional cable-stayed bridges. When using intelligent algorithms to optimize the cable force of low-tower cable-stayed bridges, due to the large stiffness of the main beam of the low-tower cable-stayed bridge, the long cable-free area near the side span and the tower root leads to uneven calculated cable force, which requires secondary cable adjustment, increasing the calculation amount of cable force optimization. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method and device, system and storage medium for optimizing the cable force of a low-tower cable-stayed bridge, so as to improve the calculation efficiency and calculation accuracy of the cable force optimization of the low-tower cable-stayed bridge.

[0004] To achieve the above object, the present invention adopts the following technical solution:

[0005] A cable force optimization method for a low-tower cable-stayed bridge, comprising:

[0006] Step S1, obtaining cable force and bending strain energy training samples of a low-tower cable-stayed bridge;

[0007] Step S2, using the IGWO algorithm to optimize the SVM parameters under the cable force and bending strain energy training samples of the low-tower cable-stayed bridge;

[0008] Step S3, taking the cable force value after the secondary tensioning as the parameter to be optimized, using the IGWO algorithm for iterative optimization, taking the cable force combination output after each generation of evolution as the input vector of SVM for nonlinear prediction, and outputting the population fitness value under the current cable force, repeatedly reducing the time cost of the optimization model for fitness evaluation, and realizing combined optimization based on IGWO.

[0009] Preferably, in step S2, the cable force and bending strain energy training samples of the low-tower cable-stayed bridge are input into the SVM model for learning, the cable force combination is used as the input vector, and the mean square error of the predicted cable force optimization value and its average cable force deviation is used to evaluate the optimization result of the gray wolf population, so as to obtain the optimal parameter combination of SVM under the IGWO algorithm selection.

[0010] Preferably, in step S3, the IGWO algorithm is used to establish a cable force optimization model for a low-tower cable-stayed bridge, and the secondary tension cable force combination T=(T1, T2, ..., T j ) is converted into the coordinates of the gray wolf individuals in the search space to form the initial population, and the fitness of the wolf pack is calculated using the predicted value of the optimized SVM to determine whether the IGWO algorithm has reached the maximum number of iterations. If so, the optimal cable force combination under the IGWO selection is output.

[0011] The present invention also provides a cable force optimization device for a low-tower cable-stayed bridge, comprising:

[0012] An acquisition module is used to obtain cable force and bending strain energy training samples of a low-tower cable-stayed bridge;

[0013] The first optimization module is used to optimize the SVM parameters under the cable force and bending strain energy training samples of the low-tower cable-stayed bridge using the IGWO algorithm;

[0014] The second optimization module is used to take the cable tension value after the secondary tensioning as the parameter to be optimized, use the IGWO algorithm for iterative optimization, use the cable tension combination output after each generation of evolution as the input vector of SVM for nonlinear prediction, and output the population fitness value under the current cable tension. This method repeatedly reduces the time cost of the optimization model for fitness evaluation and realizes combined optimization based on IGWO.

[0015] Preferably, the first optimization module inputs the cable force and bending strain energy training samples of the low-tower cable-stayed bridge into the SVM model for learning, takes the cable force combination as the input vector, and uses the mean square error of the predicted cable force optimization value and its average cable force deviation to evaluate the optimization result of the gray wolf population, and obtains the optimal parameter combination of SVM under the IGWO algorithm selection.

[0016] As a preferred method, the second optimization module uses the IGWO algorithm to establish a cable force optimization model for a low-tower cable-stayed bridge, and the secondary tension cable force combination T = (T1, T2, ..., T j ) is converted into the coordinates of the gray wolf individuals in the search space to form the initial population, and the fitness of the wolf pack is calculated using the predicted value of the optimized SVM to determine whether the IGWO algorithm has reached the maximum number of iterations. If so, the optimal cable force combination under the IGWO selection is output.

[0017] The present invention also provides a cable force optimization system for a low-tower cable-stayed bridge, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a cable force optimization method for a low-tower cable-stayed bridge when executed by the processor.

[0018] The present invention also provides a storage medium, characterized in that a computer program is stored on the storage medium, and the computer program executes the cable force optimization method of the low-tower cable-stayed bridge when running.

[0019] The present invention adopts the SVM linear prediction model optimized based on the IGWO algorithm to accurately simulate the nonlinear mapping relationship between the cable force and the structural bending strain energy of a large-span low-tower cable-stayed bridge, with a small prediction error and strong learning generalization ability. The combined optimization model established based on IGWO-SVM can effectively solve the cable force optimization problem of a low-tower cable-stayed bridge. Compared with the traditional finite element optimization method, it greatly shortens the time cost and improves the optimization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0021] Figure 1 The present invention is a flowchart of a method for optimizing the cable force of a low-tower cable-stayed bridge according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] Embodiment 1:

[0025] like Figure 1 As shown, an embodiment of the present invention provides a method for optimizing the cable force of a low-tower cable-stayed bridge, comprising:

[0026] Step S10, obtaining cable force and bending strain energy training samples of a low-tower cable-stayed bridge;

[0027] Step S20, according to the cable force and bending strain energy training samples of the low-tower cable-stayed bridge, optimizing the cable force of the low-tower cable-stayed bridge by using the IGWO-SVM algorithm; specifically comprising:

[0028] Step S201, optimizing the SVM parameters of the cable force and bending strain energy training samples of the low-tower cable-stayed bridge by using the IGWO algorithm (Improved Grey Wolf Algorithm) through the optimization model 1;

[0029] Step S202: The cable force value after the secondary tensioning is used as the parameter to be optimized through the optimization model 2, and the IGWO algorithm is used for iterative optimization. The cable force combination output after each generation of evolution is used as the input vector of the SVM for nonlinear prediction, and the fitness value of the population under the current cable force is output. This is repeated to reduce the time cost of the optimization model for fitness evaluation, and realize the combined optimization based on IGWO.

[0030] The embodiment of the present invention adopts two improvement strategies to improve the grey wolf algorithm at the same time.

[0031] 1. Chaotic map initialization population

[0032] GWO usually uses a random method to generate the initial population, but this method is difficult to maintain the population diversity of the gray wolf, and it is easy to cause the algorithm to fall into the local optimal solution when searching. In order to solve the drawbacks of randomly generating the initial population and ensure that the algorithm has a strong global search ability, the chaotic mapping method is used to initialize the population so that the initial population can be evenly distributed in the D-dimensional search space. The present invention adopts the Tent chaotic mapping method to improve the random initialization population method of the standard gray wolf algorithm. The mapping formula is shown as follows:

[0033]

[0034] Theoretically, the Tent chaotic map initialization population can generate uniformly distributed initial values ​​on [0, 1], effectively improving the global search capability and optimization speed in the early and middle stages of the algorithm, and avoiding the algorithm from falling into the local optimal solution in the early and middle stages.

[0035] 2. Cosine convergence factor

[0036] The convergence factor of GWO is determined by the distance control coefficient a. According to the expression of the distance control coefficient a, as the number of iterations increases, the control coefficient shows a linear decreasing trend in the middle and late stages of the algorithm, which seriously affects the population diversity and global search ability of the wolf pack in the late stages of the algorithm. In order to prevent the population from being unable to jump out of the local extreme value area in the middle and late stages of the algorithm, the present invention adopts a cosine-type convergence factor for improvement. The improved convergence factor value is shown in the following formula:

[0037]

[0038] Among them, a min 、a max are the minimum and maximum values ​​of the convergence factor respectively; n is the decreasing exponent; t max is the maximum number of iterations.

[0039] Optimization model 1: The values ​​of the parameters (c, g) of the SVM kernel function have a great influence on the accuracy of the SVM prediction model. Usually, the kernel function parameters are selected according to test experience, which greatly affects the prediction results. Therefore, it is necessary to optimize the values ​​of the (c, g) parameters inside the SVM. The vector machine parameters (c, g) to be optimized are converted into the coordinates of the gray wolf individuals in the search space. 62 sets of cable force combination data are randomly generated within the reasonable upper and lower limits of the designed secondary cable tension force. The data are substituted into the finite element program of the low-tower cable-stayed bridge to calculate the bending moment and deformation of the main beam and cable tower. After preprocessing the cable force-bending strain energy training samples, a sample set is formed and input into the SVM model for learning. The cable force combination is used as the input vector, and the mean square error of the predicted cable force optimization value and its average cable force deviation is used to evaluate the optimization results of the gray wolf population. Finally, the optimal parameter combination of SVM under the IGWO selection is obtained.

[0040] Optimization model 2: In order to obtain the optimal cable force combination under linear control, the IGWO algorithm is used to establish the cable force optimization model of the low-tower cable-stayed bridge. The secondary tension cable force combination T = (T1, T2, ..., T j ) is converted into the coordinates of the gray wolf individuals in the search space to form the initial population, and the fitness of the wolf pack is calculated using the predicted value of the optimized SVM to determine whether the IGWO algorithm has reached the maximum number of iterations. If so, the optimal cable force combination under the IGWO selection is output.

[0041] As an implementation method of the embodiment of the present invention, in step S20, the cable force optimization based on IGWO-SVM includes the following steps:

[0042] Step 1: Initialize the IGWO parameters of SVM parameter optimization model 1, set the gray wolf population size to n = 100, and the maximum number of iterations to t max =300, compile the parameters to be optimized (cg) into wolf group coordinates, and evenly distribute the gray wolf population into the search space according to the chaos mapping formula;

[0043]

[0044] The parameter to be optimized is the bending moment caused by adjusting the cable force of the cable-stayed bridge. The cable force adjustment value is:

[0045]

[0046] Step 2: Update the convergence factor, calculate the fitness value of the gray wolf population, and assign the optimal fitness to wolf α:

[0047]

[0048] Among them, a min 、a max are the minimum and maximum values ​​of the convergence factor respectively; n is the decreasing exponent; tmax is the maximum number of iterations.

[0049] Step 3: Determine whether the algorithm has reached the maximum number of selections. If not, return to step 2. If reached, output the optimal SVM parameter combination and use the cable force nonlinear training sample to test the SVM prediction accuracy after the updated parameters.

[0050] Step 4: Determine whether the SVM after updating the parameters meets the accuracy requirements: if not, return to step 1 to search for the best result again; if so, complete the optimization of the SVM linear prediction model;

[0051]

[0052] Step 5, initialize the IGWO parameters of optimization model 2, set the gray wolf population size: n = 100, the maximum number of selections is 300, compile the cable force combination T into wolf group coordinates, set the cable force constraints for the gray wolf population, and map the gray wolf population chaos into the search space:

[0053] U={({M L0}+[ C L]{T})} T [B]{({M L0}+[C L ]{T})}+{({M R0}+[C R ]{T})} T [B]{({M R0}+[C R ]{T})}

[0054] After optimization, the sum of bending strain energy is minimized and the value of cable force is obtained:

[0055]

[0056] The adjustment vector is a column vector composed of n (n≤m) cable forces, and its specific expression is as follows:

[0057] {T}={T1,T2,T3,…T m} T

[0058] The adjusted vector is a column vector formed by the internal force, stress and displacement of m sections at the key section position of the structure. The specific expression is as follows:

[0059] {D}={D1,D2,D3,…D m} T

[0060] Influence vector: The vector that causes the modulated vector to change when the modulating vector changes. The expression is as follows:

[0061] {C}={C 1i , C 2i , …C ii , …C mi}

[0062] Influence matrix: The matrix composed of the vector arrangement that causes the change of m modulated vectors when n modulating vectors change separately, as shown below:

[0063]

[0064] The influence matrix is ​​composed of one of the elements of mechanical quantities such as displacement, stress, internal force, etc. The structure conforms to the linear superposition principle, so it can be expressed as:

[0065] [C]{T}={D}

[0066] The influence matrix method has absorbed the advantages of other optimization methods in practice and also summarized them, so it can be optimized according to multiple objective functions in application. The optimization with bending strain energy as the optimization target is as follows:

[0067] The bending strain energy of the structure can be expressed as follows:

[0068]

[0069] Among them, M(s) is the structural bending moment function; EI is the bending stiffness of the structure.

[0070] The bridge structure can be regarded as a discrete member, and the expression is as follows:

[0071]

[0072] Where m is the total number of members; i is the member number; M Li is the bending moment at the left end of the i-th member; M Ri is the bending moment at the right end of the i-th member; L i is the length of rod i.

[0073] Rewrite formula (1) into matrix form as formula (2).

[0074] U={M L} T [B]{M L}+{M R} T [B]{M R} (2)

[0075] Among them, {M L} and {M R} are the bending moment vectors of the left and right ends of each discrete member; [B] is a diagonal matrix, and the elements on the diagonal are b i See formula

[0076]

[0077] Let the bending moment vectors at the left and right ends of the member after initial tension be {M L0} and {M R0}, the cable force adjustment value is {T}, and the bending moment vector after cable adjustment can be obtained.

[0078]

[0079] Among them, C L and C R It is the influence matrix of cable force adjustment on the left and right ends of the member.

[0080] Step 6: Update the cosine convergence factor and update the position of the wolf pack in the search space;

[0081]

[0082] D α =|C1·X α -X|

[0083] D β =|C2·X β -X|

[0084] D δ =|C3·X δ -X|

[0085] X1=X α -A1·D α

[0086] X2=X β -A2·D β

[0087] X3=X δ -A3·D δ

[0088] The final position of the ω wolf is determined by the position of the gray wolf in the first three stages, and the update formula is as follows:

[0089] X(t+1)=(X1+X2+X3) / 3

[0090] Step 7, decompile the wolf pack coordinates and use them as SVM input vectors for linear prediction, and return the predicted linear results to the optimization model 2 according to the calculation method of the objective function in formulas (3) and (4) to update the fitness value of the gray wolf population;

[0091]

[0092] After optimization, the sum of bending strain energy is minimized and the value of cable force is obtained:

[0093]

[0094] Step 8: Determine whether the algorithm has reached the termination condition. The termination condition of the algorithm is to reach the maximum number of iterations. When the algorithm has not reached the maximum number of iterations, return to step 6. When the maximum number of iterations is reached, output the coordinates of a wolf and decompile it into a cable force combination T, T = (T1, T2, ..., T j ).

[0095] The embodiment of the present invention proposes a combined optimization model of the gray wolf algorithm based on chaos mapping and nonlinear convergence improvement. By using the improved gray wolf algorithm to optimize the key parameters of the support vector machine, the best parameter combination for linear prediction under the training sample is obtained, and the optimal cable force adjustment value is further calculated to achieve the engineering goal of accurately controlling the construction of the cable-stayed bridge.

[0096] Embodiment 2:

[0097] The embodiment of the present invention further provides a cable force optimization device for a low-tower cable-stayed bridge, comprising:

[0098] An acquisition module is used to obtain cable force and bending strain energy training samples of a low-tower cable-stayed bridge;

[0099] The first optimization module is used to optimize the SVM parameters under the cable force and bending strain energy training samples of the low-tower cable-stayed bridge using the IGWO algorithm;

[0100] The second optimization module is used to take the cable tension value after the secondary tensioning as the parameter to be optimized, use the IGWO algorithm for iterative optimization, use the cable tension combination output after each generation of evolution as the input vector of SVM for nonlinear prediction, and output the population fitness value under the current cable tension. This method repeatedly reduces the time cost of the optimization model for fitness evaluation and realizes combined optimization based on IGWO.

[0101] As an implementation method of an embodiment of the present invention, the first optimization module inputs the cable force and bending strain energy training samples of a low-tower cable-stayed bridge into the SVM model for learning, takes the cable force combination as the input vector, and uses the mean square error of the predicted cable force optimization value and its average cable force deviation to evaluate the optimization result of the gray wolf population, and obtains the optimal parameter combination of SVM under the IGWO algorithm selection.

[0102] As an implementation method of the embodiment of the present invention, the second optimization module uses the IGWO algorithm to establish a cable force optimization model for a low-tower cable-stayed bridge, and the secondary tension cable force combination T=(T1, T2, ..., T j) is converted into the coordinates of the gray wolf individuals in the search space to form the initial population, and the fitness of the wolf pack is calculated using the predicted value of the optimized SVM to determine whether the IGWO algorithm has reached the maximum number of iterations. If so, the optimal cable force combination under the IGWO selection is output.

[0103] Embodiment 3:

[0104] An embodiment of the present invention further provides a cable force optimization system for a low-tower cable-stayed bridge, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a cable force optimization method for a low-tower cable-stayed bridge when executed by the processor.

[0105] Embodiment 4:

[0106] An embodiment of the present invention further provides a storage medium having a computer program stored thereon, and the computer program executes a method for optimizing the cable force of a low-tower cable-stayed bridge when the computer program is run.

[0107] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for optimizing cable force of a low-tower cable-stayed bridge, characterized in that: include: Step S1, obtaining cable force and bending strain energy training samples of a low-tower cable-stayed bridge; Step S2, using the IGWO algorithm to optimize the SVM parameters under the cable force and bending strain energy training samples of the low-tower cable-stayed bridge; Step S3, taking the cable force value after the secondary tensioning as the parameter to be optimized, using the IGWO algorithm for iterative optimization, taking the cable force combination output after each generation of evolution as the input vector of SVM for nonlinear prediction, and outputting the population fitness value under the current cable force, repeatedly reducing the time cost of the optimization model for fitness evaluation, and realizing combined optimization based on IGWO.

2. The cable force optimization method for a low-tower cable-stayed bridge according to claim 1, characterized in that: In step S2, the cable force and bending strain energy training samples of the low-tower cable-stayed bridge are input into the SVM model for learning. The cable force combination is used as the input vector, and the mean square error of the predicted cable force optimization value and its average cable force deviation is used to evaluate the optimization result of the gray wolf population, and the optimal parameter combination of SVM under the IGWO algorithm selection is obtained.

3. The cable force optimization method for a low-tower cable-stayed bridge according to claim 2, characterized in that: In step S3, the IGWO algorithm is used to establish a cable force optimization model for a low-tower cable-stayed bridge. The secondary tension cable force combination is converted into the coordinates of the gray wolf individuals in the search space to form an initial population. The fitness of the wolf pack is calculated using the predicted value of the optimized SVM to determine whether the IGWO algorithm has reached the maximum number of iterations. If so, the optimal cable force combination under the IGWO selection is output.

4. A cable force optimization device for a low-tower cable-stayed bridge, characterized in that: include: An acquisition module is used to obtain cable force and bending strain energy training samples of a low-tower cable-stayed bridge; The first optimization module is used to optimize the SVM parameters under the cable force and bending strain energy training samples of the low-tower cable-stayed bridge using the IGWO algorithm; The second optimization module is used to take the cable tension value after the secondary tensioning as the parameter to be optimized, use the IGWO algorithm for iterative optimization, use the cable tension combination output after each generation of evolution as the input vector of SVM for nonlinear prediction, and output the population fitness value under the current cable tension. This method repeatedly reduces the time cost of the optimization model for fitness evaluation and realizes combined optimization based on IGWO.

5. The cable force optimization device for a low-tower cable-stayed bridge according to claim 4, characterized in that: The first optimization module inputs the cable force and bending strain energy training samples of the low-tower cable-stayed bridge into the SVM model for learning. The cable force combination is used as the input vector, and the mean square error of the predicted cable force optimization value and its average cable force deviation is used to evaluate the optimization results of the gray wolf population, and the optimal parameter combination of SVM under the IGWO algorithm selection is obtained.

6. The cable force optimization device for a low-tower cable-stayed bridge according to claim 5, characterized in that: The second optimization module uses the IGWO algorithm to establish a cable force optimization model for low-tower cable-stayed bridges, converts the secondary tension cable force combination into the coordinates of gray wolf individuals in the search space to form an initial population, and uses the predicted value of the optimized SVM to calculate the fitness of the wolf pack to determine whether the IGWO algorithm has reached the maximum number of iterations. If so, the optimal cable force combination under the IGWO selection is output.

7. A cable force optimization system for a low-tower cable-stayed bridge, characterized in that: include: A memory and a processor, wherein the memory stores a computer program executed by the processor, and when the computer program is executed by the processor, the method for optimizing the cable force of a low-tower cable-stayed bridge as described in any one of claims 1 to 3 is executed.

8. A storage medium, characterized in that: The storage medium stores a computer program, which, when running, executes the cable force optimization method for a low-tower cable-stayed bridge according to any one of claims 1 to 3.

Images