Determination method for internal force of suspender, electronic equipment and storage medium

Abstract

The invention discloses a boom internal force determination method, electronic equipment and a storage medium, and the method comprises the steps: obtaining an initial internal force data set corresponding to a boom of a target bridge, and taking the initial internal force data set as a current internal force data set; the current internal force data set and the structural information of the target bridge are input into a bridge response prediction model, a current response data set is obtained, and the bridge response prediction model is a model used for simulating a finite element model of the target bridge; if an iteration termination condition is satisfied, determining the current internal force data in the current internal force data set as the suspender internal force of the target bridge; if the iteration termination condition is not met, determining a next internal force data set according to the current internal force data set, the current response data set and the bridge response prediction model, and taking the next internal force data set as the current internal force data set; and returning to the step of inputting the current internal force data set and the structural information of the target bridge into the pre-trained bridge response prediction model. According to the method, the internal force of the arch bridge suspender is optimized.

Description

Technical Field

[0001] The present invention relates to the field of structural design, and in particular, to a method for determining the internal force of a suspension rod, an electronic device, and a storage medium. Background Art

[0002] When designing an arch bridge such as a reticulated suspension rod tied arch bridge, since the internal force of the suspension rod plays an important role in the structural state of the arch bridge, therefore, the optimization of the internal force of the suspension rod in the completed bridge state is the key point in the design process of the arch bridge.

[0003] Currently, in the process of optimizing the internal force of the suspension rod, a finite element model of the arch bridge is established, and mechanical analysis of the arch bridge is carried out by using methods such as the zero displacement method or the rigid support method to ensure the safety and reliability of the arch bridge design after the internal force of the suspension rod is optimized. However, when optimizing the internal force of the suspension rod of a large-scale arch bridge based on the finite element model and the above methods, due to factors such as a large number of suspension rods of the arch bridge and a large number of design internal force selections for different suspension rods, the computing power and time cost required for the optimization process based on the finite element model are relatively large. Summary of the Invention

[0004] The present invention provides a method for determining the internal force of a suspension rod, an electronic device, and a storage medium. This method solves the problem that only using a finite element model to optimize the internal force of the suspension rod currently will result in relatively large computing power and time cost in the optimization process. It can not only optimize large-scale arch bridges, reduce computing power and time cost, but also quickly and accurately determine the mechanical response of the bridge during the process of optimizing the internal force of the suspension rod.

[0005] According to one aspect of the present invention, a method for determining the internal force of a suspension rod is provided, and the method includes:

[0006] Obtain an initial internal force data set corresponding to the suspension rod of the target bridge, and use the initial internal force data set as the current internal force data set.

[0007] Input the current internal force data set and the structural information of the target bridge into a pre-trained bridge response prediction model to obtain a current response data set, where the bridge response prediction model is a model trained based on a neural network and used to simulate the finite element model of the target bridge, and the finite element model of the target bridge is a simulation model determined based on the design requirements and design scheme of the target bridge.

[0008] If the iteration termination condition is satisfied, determine the current internal force data in the current internal force data set as the internal force of the suspension rod of the target bridge.

[0009] If the iteration termination condition is not satisfied, based on the current internal force data set, the current response data set, and the bridge response prediction model, determine the next internal force data set, and use the next internal force data set as the current internal force data set, and return to execute the step of inputting the current internal force data set and the structural information of the target bridge into the pre-trained bridge response prediction model to obtain the current response data set.

[0010] The method for determining the internal force of the suspender provided by the embodiment of the present invention includes: obtaining the initial internal force data set corresponding to the suspender of the target bridge, and using the initial internal force data set as the current internal force data set; inputting the current internal force data set and the structural information of the target bridge into the pre-trained bridge response prediction model to obtain the current response data set; if the iteration termination condition is satisfied, determine the current internal force data in the current internal force data set as the internal force of the suspender of the target bridge; if the iteration termination condition is not satisfied, based on the current internal force data set, the current response data set, and the bridge response prediction model, determine the next internal force data set, and use the next internal force data set as the current internal force data set, and return to execute the step of inputting the current internal force data set and the structural information of the target bridge into the pre-trained bridge response prediction model to obtain the current response data set. In the above technical solution, on the one hand, using the trained bridge response prediction model to replace the finite element model realizes the rapid and accurate determination of the mechanical response of the target bridge when the suspender of the target bridge is designed according to each internal force data in the current internal force data set, providing accurate data support for determining the internal force of the suspender of the target bridge. On the other hand, based on the iteration termination condition, it is judged whether the current internal force data in the current internal force data set can be determined as the internal force of the suspender of the target bridge; when the iteration termination condition is not satisfied, by updating the current internal force data set, different suspender design internal forces are traversed, and the mechanical response of the bridge under different suspender design internal forces is determined, solving the problem that the current optimization of the suspender internal force only using the finite element model will result in a large computational power and time cost in the optimization process. It can not only optimize large-scale arch bridges, reduce the computational power and time cost, but also quickly and accurately determine the mechanical response of the bridge during the optimization of the suspender internal force; when the iteration termination condition is satisfied, the current internal force data is determined as the internal force of the suspender of the target bridge, realizing the traversal of different suspender design internal forces and ensuring the superiority of the finally determined suspender internal force.

[0011] According to another aspect of the present invention, there is provided a device for determining the internal force of a suspender, the device includes:

[0012] An acquisition module, configured to acquire an initial internal force data set corresponding to the suspender of the target bridge, and use the initial internal force data set as the current internal force data set.

[0013] An output module, configured to input the current internal force dataset and the structural information of the target bridge into a pre-trained bridge response prediction model to obtain a current response dataset. The bridge response prediction model is a model trained based on a neural network and used to simulate the finite element model of the target bridge, and the finite element model of the target bridge is a simulation model determined based on the design requirements and design scheme of the target bridge.

[0014] A judgment module, configured to, if the iteration termination condition is satisfied, determine the current internal force data in the current internal force dataset as the hanger internal force of the target bridge; if the iteration termination condition is not satisfied, determine a next internal force dataset according to the current internal force dataset, the current response dataset, and the bridge response prediction model, and use the next internal force dataset as the current internal force dataset, and return to execute the step of inputting the current internal force dataset and the structural information of the target bridge into the pre-trained bridge response prediction model to obtain the current response dataset.

[0015] According to another aspect of the present invention, there is provided an electronic device, which includes:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program executable by the at least one processor. The computer program is executed by the at least one processor so that the at least one processor can execute the method for determining the hanger internal force according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, there is provided a computer-readable storage medium storing computer instructions for causing a processor to implement the method for determining the hanger internal force according to any embodiment of the present invention when executed.

[0020] According to another aspect of the present invention, there is provided a computer program product including a computer program, and the computer program implements the method for determining the hanger internal force according to any embodiment of the present invention when executed by a processor.

[0021] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. Description of the Drawings

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for description in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0023] Figure 1 It is a schematic flowchart of a method for determining the internal force of a suspender provided by an embodiment of the present invention;

[0024] Figure 2 It is a schematic flowchart of another method for determining the internal force of a suspender provided by an embodiment of the present invention;

[0025] Figure 3 It is an example diagram of the training process of a bridge response prediction model provided by an embodiment of the present invention;

[0026] Figure 4 It is a schematic structural diagram of a device for determining the internal force of a suspender provided by an embodiment of the present invention;

[0027] Figure 5 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0028] In order to enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0029] It should be noted that the terms "current", "next", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order different from those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily need to be limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0030] Figure 1The figure is a schematic flowchart of a method for determining the internal force of a suspender provided by an embodiment of the present invention. This embodiment is applicable to the situation of determining the optimal internal force of the suspender when optimizing the internal force of the suspender of a tied-arch bridge. This method can be executed by a device for determining the internal force of the suspender. The device for determining the internal force of the suspender can be implemented in the form of hardware and / or software, and the device for determining the internal force of the suspender can be configured in an electronic device. In this embodiment, the electronic device can be a computer or a terminal device. As Figure 1 shown, the method includes:

[0031] S101. Obtain an initial internal force data set corresponding to the suspenders of the target bridge, and use the initial internal force data set as the current internal force data set.

[0032] Among them, the target bridge is the bridge to be designed that needs to optimize the internal force of the suspender this time. In this embodiment, the target bridge can be a tied-arch bridge with suspenders, such as a reticulated tied-arch bridge and a vertical tied-arch bridge, etc. The initial internal force data set is the first obtained internal force data set before executing this method. In this embodiment, a single internal force data set includes multiple internal force data, and a single internal force data represents the internal force designed for all the suspenders on the target bridge. That is, since there may be multiple suspenders on a target bridge, a single internal force data includes multiple internal force values, and a single internal force value represents the internal force value designed for a single suspender on the target bridge. For example, if there are 6 suspenders on the target bridge, a single internal force data can be T 1 = [Ta, Tb, Tc, Td, Te, Tf], where each internal force value corresponds to the internal force of a single suspender. For a single internal force data set, it can be T = {T 1 , T 2 , T 3 , …}. Among them, at least one of the internal force values in T 2 is different from the internal force value at the corresponding position in T 1 , and the same applies to other internal force data. The current internal force data set is the internal force data set for executing the current loop iteration after entering the following steps.

[0033] Specifically, for a target bridge, before design, its design requirements, design specifications, and basic information will be obtained. For example, its span, bridge width, and the number of suspenders will be obtained. Therefore, an initial internal force data set can be directly generated based on this information of the target bridge. For example, according to the above information of the target bridge and the design constraints, the constraint conditions for the internal force of the suspender can be determined, and the initial internal force data set can be generated based on the constraint conditions for the internal force of the suspender. After generating the initial internal force data set, the initial internal force data set can be used as the current internal force data set, and subsequent steps can be prepared for execution.

[0034] Optionally, the number of suspenders for each arch rib of the target bridge is denoted as 2v, which is divided into two groups along the longitudinal and transverse directions of the bridge. The two groups of suspenders are symmetrically arranged about the longitudinal center axis of the arch rib. The internal force distributions of the two groups of suspenders are symmetrically equal about the longitudinal center axis of the arch bridge. An internal force data of a suspender is T * =[Ta, Tb, …, Tq, …, Tv], where “*” represents any number, such as 1, 2, 3, ….

[0035] S102. Input the current internal force dataset and the structural information of the target bridge into a pre-trained bridge response prediction model to obtain the current response dataset.

[0036] Among them, the bridge response prediction model is a model based on neural network training and used to simulate the finite element model of the target bridge. The finite element model of the target bridge is a simulation model determined based on the design requirements and design scheme of the target bridge. The response dataset includes multiple response data, and each response data corresponds to an internal force data in the internal force dataset. The structural information is the relevant information of the structure of the target bridge. In this embodiment, the structural information includes: the geometric information of the tie beam, arch rib elements, etc. The response data is the mechanical response data of each structure of the target bridge after the internal force design of the suspenders is carried out according to the corresponding internal force data. In this embodiment, the response data includes: the moment of the tie beam element, the moment of the arch rib element, the stress of the arch foot element, and the deflection of the mid-span element of the tie beam, etc.

[0037] Specifically, training the bridge response prediction model can be used to replace using the finite element model to determine the mechanical response data of the target bridge after the internal force optimization of the suspenders. Therefore, after obtaining the current internal force dataset, the current internal force dataset and the structural information of the target bridge can be input into the bridge response prediction model, and the bridge response prediction model will output the current response dataset corresponding to the current internal force dataset.

[0038] Among them, for the training method of the bridge response prediction model, in one implementation, it is directly trained into a general model that can be applied to all bridges. In another implementation, since the structures of each bridge are different, different design schemes will also lead to different mechanical responses. Therefore, in order to ensure that the output of the bridge response prediction model more accurately reflects the mechanical response of the current target bridge, after each target bridge is determined, a bridge response prediction model suitable for the target bridge can be trained first, and then this model is used to replace the finite element model.

[0039] In this embodiment, the trained bridge response prediction model is used to replace the finite element model. The current internal force dataset designed for the suspenders of the target bridge is used as the input of the model, and the corresponding current response dataset of the current internal force dataset is output, realizing the rapid and accurate determination of the mechanical response of the target bridge after the suspenders of the target bridge are designed according to each internal force data in the current internal force dataset, providing accurate data support for determining the internal force of the suspenders of the target bridge later.

[0040] S103. Determine whether the iteration termination condition is satisfied; if the iteration termination condition is satisfied, execute S104; if the iteration termination condition is not satisfied, execute S105.

[0041] Among them, the iteration termination condition is preset. In this embodiment, the iteration termination condition can be: the number of iterations of the current internal force dataset exceeds the preset number of times or the fitness corresponding to the data in the current internal force dataset meets the preset fitness requirement. For example, the difference in fitness corresponding to each data in the current internal force dataset and the fitness corresponding to each data in the internal force dataset in the historical process is less than the preset difference, or the difference between the fitness corresponding to each data in the current internal force dataset is small. In this embodiment, if the initial internal force dataset is directly used as the current internal force dataset, the number of iterations of this current internal force dataset is 0.

[0042] Specifically, determine whether the situation corresponding to the current internal force dataset meets the iteration termination condition. If the iteration termination condition is met, it means that there is no need to perform the next round of iteration. Therefore, S104 can be directly executed. If the iteration termination condition is not met, it means that the next round of iteration is still required. Therefore, S105 can be directly executed.

[0043] S104. Determine the current internal force data in the current internal force dataset as the internal force of the suspenders of the target bridge.

[0044] Specifically, if the iteration termination condition is met, the current internal force data in the current internal force dataset can be directly determined as the internal force of the suspenders of the target bridge. For example, the current internal force data with the optimal fitness corresponding to the current internal force data in the current internal force dataset can be directly determined as the internal force of the suspenders of the target bridge.

[0045] S105. Determine the next internal force dataset according to the current internal force dataset, the current response dataset and the bridge response prediction model, and use the next internal force dataset as the current internal force dataset; return to execute S102.

[0046] Among them, the next internal force dataset is the internal force dataset after updating the current internal force dataset. In this embodiment, after determining the next internal force dataset, the number of iterations is incremented by one.

[0047] Specifically, if the iteration termination condition is not satisfied, the current internal force dataset can be updated. Therefore, the current internal force dataset can be updated based on the current internal force dataset, the current response dataset, and the bridge response prediction model to obtain the next internal force dataset.

[0048] Exemplarily, in one implementation, first determine the candidate internal force data based on the current internal force data and the current response data corresponding to the current internal force data, perform an optimization and expansion process on the candidate internal force data, and use the data after the optimization and expansion process as the internal force data in the next internal force dataset. Among them, the optimization and expansion processing methods include: simulated annealing algorithm, particle swarm optimization algorithm, ant colony algorithm, genetic algorithm, etc. In another implementation, methods such as random generation, data fitting, machine learning algorithms, etc. can be used to determine new internal force data based on the current internal force data in the current internal force dataset, then input the new internal force data into the bridge response prediction model to obtain the response data corresponding to the new internal force data, and then determine the next internal force dataset based on the new internal force data, the response data corresponding to the new internal force data, the current internal force data, and the current response data corresponding to the current internal force data.

[0049] In this embodiment, based on the iteration termination condition, it is judged whether the current internal force data in the current internal force dataset can be determined as the hanger internal force of the target bridge; when the iteration termination condition is not satisfied, by updating the current internal force dataset, different hanger design internal forces are traversed, and the mechanical response of the bridge under different hanger design internal forces is determined, solving the problem that only using the finite element model to optimize the hanger internal force currently will result in a large computational power and time cost in the optimization process. It can not only reduce the computational power and time cost when optimizing a large-scale arch bridge, but also quickly and accurately determine the mechanical response of the bridge during the process of optimizing the hanger internal force; when the iteration termination condition is satisfied, the current internal force data is determined as the hanger internal force of the target bridge, realizing the traversal of different hanger design internal forces and ensuring the superiority of the hanger internal force.

[0050] The method for determining the internal force of the suspender provided by the embodiment of the present invention includes: obtaining the initial internal force data set corresponding to the suspender of the target bridge, and using the initial internal force data set as the current internal force data set; inputting the current internal force data set and the structural information of the target bridge into the pre-trained bridge response prediction model to obtain the current response data set; if the iteration termination condition is satisfied, determining the current internal force data in the current internal force data set as the internal force of the suspender of the target bridge; if the iteration termination condition is not satisfied, determining the next internal force data set according to the current internal force data set, the current response data set and the bridge response prediction model, and using the next internal force data set as the current internal force data set, and returning to execute the step of inputting the current internal force data set and the structural information of the target bridge into the pre-trained bridge response prediction model to obtain the current response data set. In the above technical solution, on the one hand, the trained bridge response prediction model is used to replace the finite element model, and the current internal force data set designed for the suspender of the target bridge is used as the input of the model, and the current response data set corresponding to the current internal force data set is output, realizing the rapid and accurate determination of the mechanical response of the target bridge after the suspender of the target bridge is designed according to each internal force data in the current internal force data set, providing accurate data support for determining the internal force of the suspender of the target bridge. On the other hand, based on the iteration termination condition, it is judged whether the current internal force data in the current internal force data set can be determined as the internal force of the suspender of the target bridge; when the iteration termination condition is not satisfied, by updating the current internal force data set, different suspender design internal forces are traversed, and the mechanical response of the bridge under different suspender design internal forces is determined, solving the problem that the current optimization of the suspender internal force only using the finite element model will lead to large computing power and time costs required for the optimization process. It can not only optimize large-scale arch bridges, reduce computing power and time costs, but also quickly and accurately determine the mechanical response of the bridge during the process of optimizing the suspender internal force; when the iteration termination condition is satisfied, the current internal force data is determined as the internal force of the suspender of the target bridge, realizing the traversal of different suspender design internal forces and ensuring the superiority of the finally determined suspender internal force.

[0051] Figure 2 FIG. is a schematic flow chart of another method for determining the internal force of the suspender provided by the embodiment of the present invention. On the basis of the above embodiment and other examples, the steps after the iteration termination condition is not satisfied are described in detail.

[0052] Prior to this, an exemplary description of the training method of the bridge response prediction model provided by the embodiment of the present invention is given first. Figure 3 FIG. is an example diagram of the training process of the bridge response prediction model provided by the embodiment of the present invention. As Figure 3 shown, the method includes:

[0053] S301. Obtain the sample internal force dataset, and input the sample internal force dataset into the finite element model to obtain the actual response dataset.

[0054] Among them, the sample internal force dataset is the input sample used to train the bridge response prediction model. The actual response dataset is the actual response dataset of the target bridge corresponding to the sample internal force dataset determined by the finite element model based on the sample internal force dataset. In this embodiment, the structural dimensions of the arch bridge, the material parameters of each component, and the load conditions are known conditions. The tie beam and arch rib can be simulated using beam elements, and the suspenders can be simulated using rod elements.

[0055] Specifically, the finite element model of the arch bridge can be established first based on the design scheme and design requirements of the arch bridge. For example, the design scheme and design requirements of the arch bridge are as follows: the span is 152 meters, the bridge width is 29.7 meters, the arch bridge is provided with two arch ribs, the arch axis is a parabola, the arch rib rise is 26 meters, and the rise-span ratio is 0.17. Among them, the arch ribs, tie beams, wind braces, cross braces, and small longitudinal beams of the arch bridge are all Q345 steel box girders. The arch bridge has 72 suspenders, 36 suspenders for each arch rib, the cross-sectional area of the suspenders is 21.7 square centimeters, the average included angle between the suspenders and the tie beam is 63.5 degrees, the spacing between the same-direction suspenders is 7.5 meters, and the tensile strength is 1770 MPa. The arch ribs, tie beams, wind braces, cross braces, and small longitudinal beams all adopt beam elements, and the suspenders adopt rod elements. At this time, the finite element model of the target bridge can be established based on the above information.

[0056] Secondly, since there are preset design parameters determined in advance during the design process of the bridge. For example, the stress of the arch foot unit is less than the design value of the material strength used, the deflection of the tie beam mid-span unit is less than 1 / 600 of the tie beam span, and the internal force of each suspender is less than 0.4 times the ultimate tensile strength of the material used. Therefore, the internal force constraint conditions of the suspenders can be determined based on the preset design parameters and the finite element model. For example, continuing with the above example, it can be determined that the internal force of the suspenders should be less than 1699.2 kN. Based on the internal force constraint conditions of the suspenders, the initial internal force dataset can be randomly generated. For example, the sample internal force dataset can be randomly generated based on statistical distribution methods, Monte Carlo simulation methods, random perturbation methods, and random sampling methods.

[0057] Finally, input the sample internal force dataset into the finite element model, and the actual response dataset corresponding to the sample internal force dataset can be obtained. At this time, the actual response dataset can be used as the "label" for training the model.

[0058] S302. Input the sample internal force dataset and the structural information into the neural network to obtain the sample response dataset.

[0059] Among them, the input data of the model in this embodiment, that is, the data in the internal force dataset, is the internal force of the suspender of the target bridge, and the structural information includes the geometric information of the crossbeam and the arch rib unit. The output data of the model, that is, the response dataset, includes the bending moment of the crossbeam unit, the bending moment of the arch rib unit, the stress of the arch foot unit, and the deflection of the mid-span unit of the crossbeam.

[0060] Specifically, first determine the network structure and parameters of the neural network, and initialize the weights and biases of each neuron. Then, the sample internal force dataset and the structural information of the target bridge can be input into the neural network to obtain the sample response dataset corresponding to the sample internal force dataset.

[0061] Optionally, the network structure and parameters of the neural network can be: a BP neural network with 2 hidden layers, and each hidden layer contains 100 neurons. The activation function used can be the logistic (Sigmoid) function, as shown in the formula: sigmoid(x) = 1 / (1 + e (-x) ). Among them, sigmoid(·) represents the output of the neuron, x is the input of the neuron, and e is the natural constant.

[0062] Optionally, before inputting the sample internal force dataset into the neural network, for the sake of ensuring the normalization of variables, a normalization process can be carried out, and the normalization formula is as follows:

[0063]

[0064] Among them, X is each variable, X non represents the normalized variable, min(X) is the minimum value of the variable, and max(X) is the maximum value of the variable. In this embodiment, "variable" not only refers to the data in the sample internal force dataset, but also can refer to the structural information of the target bridge input into the model and the actual response dataset corresponding to the sample internal force dataset. That is, all data related to training the model, as long as it is a variable, can be normalized to ensure the accuracy and normalization of model training.

[0065] Optionally, before inputting the sample internal force dataset into the neural network, the internal force data in the sample internal force dataset can also be divided into a training set, a validation set, and a test set according to a ratio of 70%, 15%, and 15%, and the model can be trained, validated, and tested.

[0066] Optionally, in determining the network structure and parameters, the number of network hidden layers, the number of nodes in the hidden layer, etc. are usually obtained based on experience. The number of nodes in the hidden layer can refer to the following empirical formula:

[0067]

[0068] Among them, n h represents the number of nodes in the hidden layer, ni Represents the number of input variables, n o Represents the number of output variables, and a is a constant from 2 to 10. One of the above three expressions can be selected according to requirements.

[0069] Optionally, initialize the weights and biases of each neuron, which can be done according to the following general formula:

[0070]

[0071] Among them, y j is the output of neuron j, f(·) is the activation function, x i is the i-th input of neuron j, w i,j is the weight of the i-th input of neuron j, b j is the bias term of neuron j, and n is the total number of inputs to neuron j. In this embodiment, the above Sigmoid function is only the formula of a specific method under the general formula.

[0072] S303. Determine the training error based on the sample response dataset and the actual response dataset, and determine whether the training error meets the preset error; if so, execute S304; if not, execute S305.

[0073] Specifically, since the actual response dataset is obtained based on the finite element model, the actual response dataset can be used as the standard data to determine the accuracy of the sample response dataset obtained based on the model. Therefore, the training error can be determined based on the sample response dataset and the actual response dataset, and the training error can be used as a performance evaluation index. Compare the training error with the preset error. If it meets the preset error, for example, is less than or equal to the preset error, then execute S304. If it does not meet the preset error, for example, is greater than the preset error, then execute S305.

[0074] Exemplarily, the mean square error (MSE) between the sample response dataset and the actual response dataset can be calculated. Optionally, the conjugate gradient descent method is selected as the network training algorithm. The termination condition for model training can also be: the maximum number of training epochs Epoch is 1500, and the training termination MSE is set to 1‰. That is, the preset error is that MSE is less than or equal to 1‰. Also, following the above example, the model can be trained using the training set until the maximum number of training epochs is reached. The validation set can be used to determine the training error and determine whether the training error meets the preset error to verify the training effect of the model. The test set can also be used for testing to further determine the training effect of the model.

[0075] S304. Determine the neural network as the bridge response prediction model.

[0076] Specifically, if the training error meets the preset error, it can be determined that the training of the neural network ends. At this time, the neural network can be determined as the bridge response prediction model.

[0077] S305. Update the model parameters of the neural network; return to execute S302.

[0078] Specifically, if the training error does not meet the preset error, the model parameters of the neural network need to be updated. The specific update method can be the method of updating the model parameters of the neural network during the conventional model training process, which will not be elaborated here. After the update, it is necessary to return to execute the step of inputting the sample internal force dataset and the structural information into the neural network to obtain the sample response dataset.

[0079] Optionally, the weight and bias terms can be updated iteratively until the convergence condition of network training is reached.

[0080] In this embodiment, first, an actual response dataset of the sample internal force dataset for the target bridge is generated based on the finite element model, providing standard output data for training the bridge response prediction model that can replace the finite element model later. Then, the neural network is trained using the sample internal force dataset and the structural information of the target bridge, and the training error is determined based on the actual response dataset and the sample response dataset to detect the accuracy of the trained neural network. Also, when the training does not meet the stop requirement, the model parameters of the neural network are updated to solve the problem that when processing the internal force data of a large number of suspenders based on the finite element model, it will result in a large amount of computing power and large time cost.

[0081] Continue to refer to Figure 2 As Figure 2 shown, the method includes:

[0082] S201. Determine the internal force constraint conditions of the suspenders according to the finite element model and the preset design parameters.

[0083] Specifically, the preset design parameters are some pre-set design parameter limitations, such as the arch springing unit stress being less than the design value of the material strength used, the mid-span unit deflection of the crossbeam being less than 1 / 600 of the crossbeam span, and the internal force of each suspender being less than 0.4 times the ultimate tensile strength of the material used. Using the preset design parameters and based on the simulation structure of the target bridge shown by the finite element model, the internal force constraint conditions of the suspenders can be determined.

[0084] S202. Randomly generate an initial internal force dataset according to the internal force constraint conditions of the suspenders.

[0085] Specifically, the internal force constraint conditions of the suspenders can be used to limit the internal force data in the generated internal force dataset. Therefore, the initial internal force dataset can be randomly generated based on methods such as the Monte Carlo sampling method or the Latin hypercube sampling method.

[0086] In this embodiment, based on the method of random generation, an initial internal force dataset is generated, which can provide diverse and random data for determining the internal force of the hanger with the optimized internal force in the subsequent process, and improve the robustness of the hanger internal force.

[0087] S203. Use the initial internal force dataset as the current internal force dataset.

[0088] Specifically, after obtaining the initial internal force dataset, the initial internal force dataset can be used as the current internal force dataset.

[0089] S204. Input the current internal force dataset and the structural information of the target bridge into the pre-trained bridge response prediction model to obtain the current response dataset.

[0090] Specifically, the current internal force dataset and the structural information of the target bridge can be input into the bridge response prediction model, and the bridge response prediction model will output the current response dataset corresponding to the current internal force dataset.

[0091] Optionally, in this embodiment, the hanger response model is only used for the current target bridge. If the target bridge changes, or the overall structure of the target bridge remains unchanged but the structural data information changes, in order to ensure the accuracy of the final result, the model needs to be retrained. For example, if the bridge type is completely different, the model needs to be retrained. However, if only some basic parameters are changed (such as the cross-section parameters of the arch rib and tie beam, the span of the bridge, etc.), parametric modeling can be used to enhance the coverage of the dataset.

[0092] S205. Determine whether the iteration termination condition is satisfied; if so, execute S206; if not, execute S207.

[0093] Specifically, in this embodiment, the iteration termination condition includes at least one of the following:

[0094] The fitness corresponding to the data in the current internal force dataset meets the preset fitness requirement; the number of iterations of the current internal force dataset exceeds the preset number of times.

[0095] Among them, the calculation method of the fitness is determined according to the current internal force dataset and the current response dataset, and the calculation method in the following steps can be referred to, which will not be elaborated here. The preset fitness requirement may include, for example: the difference between the fitness corresponding to the data in the current internal force dataset and the fitness corresponding to the data in the historical internal force dataset is less than the preset difference, or the difference between the fitness corresponding to the data in the current internal force dataset is relatively small.

[0096] Specifically, when obtaining the current response dataset, it can be determined whether the iteration termination condition is satisfied. If the iteration termination condition is satisfied, S206 can be directly executed. If the iteration termination condition is not satisfied, S207 needs to be executed.

[0097] Optionally, the iteration termination condition in this embodiment can be one or more of the above conditions, and other conditions can also be used as the iteration termination condition.

[0098] S206. Determine the current internal force data in the current internal force dataset as the hanger internal force of the target bridge.

[0099] Specifically, if the iteration termination condition is satisfied, directly determine the current internal force data in the current internal force dataset as the hanger internal force of the target bridge.

[0100] Exemplarily, if there are multiple current internal force data in the current internal force dataset; optionally, select the internal force data with the smallest fitness corresponding to the current internal force data as the hanger internal force of the target bridge; optionally, it can be selected by the user himself.

[0101] In this embodiment, if the iteration termination condition is satisfied, the current internal force data in the current internal force dataset can be directly determined as the hanger internal force of the target bridge, realizing the fast and accurate determination of the preferred internal force data as the hanger internal force from the diverse hanger internal force datasets, ensuring less computing power and lower time cost in the optimization process of the hanger internal force of the target bridge.

[0102] S207. Determine the candidate internal force data and its corresponding candidate response data according to the current internal force dataset, the current response dataset, and the bridge response prediction model.

[0103] Among them, the candidate internal force data is the candidate internal force data determined for the next round of optimization after updating and optimizing the current internal force dataset.

[0104] Specifically, if the iteration termination condition is not satisfied, the current internal force dataset needs to be updated and optimized. At this time, the update and optimization can be carried out according to the following steps:

[0105] (1) Obtain the preset initialization control parameters.

[0106] Among them, the initialization control parameters include population size, crossover rate, mutation rate, number of iterations, and termination condition.

[0107] Specifically, before performing genetic iteration optimization, the preset initialization control parameters will be determined first.

[0108] Exemplarily, in this example, the population size is set to 100, the crossover rate is set to 0.8, the mutation rate is set to 0.1, and the number of iteration terminations is set to 1000 times.

[0109] (2) Determine the intermediate internal force dataset according to the initialization control parameters, the internal force constraint conditions of the suspenders, and the current internal force dataset.

[0110] Specifically, determining the intermediate internal force dataset is the step of initializing the parental population. For example, use the current internal force data in the current internal force dataset as the individual variables of the population. Among them, the constraint condition is the internal force constraint condition of the suspenders.

[0111] Optionally, binary encoding can also be performed on the input variables of the population, genetic operators such as crossover and mutation are used on the binary variables to generate a new generation of population (offspring population), and decoding operations are performed on the variables of the new generation of population. And, repeat the above steps, screen and update each generation of population until the genetic termination condition is reached, and finally use the last new generation of population obtained as the intermediate internal force dataset.

[0112] (3) Input the intermediate internal force dataset and the structural information into the bridge response prediction model to obtain the intermediate response dataset.

[0113] Specifically, input the newly generated intermediate internal force dataset and the structural information into the bridge response prediction model, and the intermediate response dataset corresponding to the intermediate internal force dataset can be obtained.

[0114] (4) Determine each data in the current internal force dataset and the intermediate internal force dataset as candidate internal force data; and determine each data in the current response dataset and the intermediate response dataset as candidate response data.

[0115] Specifically, after obtaining the intermediate internal force dataset and the intermediate response dataset, it can be determined that the data update has been completed based on the current internal force dataset. Therefore, the data in the obtained current internal force dataset and the intermediate internal force dataset can be merged, and all the internal force data are used as candidate internal force data. Correspondingly, the response data corresponding to each internal force data is the candidate response data.

[0116] S208. For each candidate internal force data and its corresponding candidate response data, determine the candidate fitness corresponding to each candidate internal force data according to the candidate internal force data and its corresponding candidate response data.

[0117] Specifically, the method for determining the candidate fitness is as follows:

[0118] (1) Determine the optimization constraint limit according to the internal force constraint condition of the suspender and the candidate response data.

[0119] Specifically, the internal force constraint condition of the suspender is used to constrain the internal force of the suspender. The candidate response data and the preset design parameters can determine the mechanical response constraints of other structures of the target bridge. Therefore, the optimization constraint limit can be finally determined based on the above content.

[0120] Among them, the optimization constraints only need to be determined in the first calculation and do not need to be determined every time during the loop process.

[0121] (2) Taking the preset strain information of the target bridge as the optimization objective, the optimization constraints as the optimization constraint conditions, and the candidate internal force data as the optimization variables, the candidate fitness is determined.

[0122] Among them, the preset strain information is the pre-determined optimization objective required by the user. For example, the minimum bending strain energy of the arch rib and the tie beam can be the preset strain information. The candidate fitness is the objective function of the optimization problem.

[0123] Specifically, the preset strain information of the target bridge can be used as the optimization objective of the optimization problem, the optimization constraints as the optimization constraint conditions of the optimization problem, and the candidate internal force data as the optimization variables of the optimization problem to determine the objective function of the optimization problem, that is, the candidate fitness.

[0124] Exemplarily, taking the solution of the general fitness as an example, it is as follows:

[0125] Continuing the above example, according to the preset design parameters, the finite element model, and the structural data of the target bridge and other information, it can be determined respectively that the stress of the arch foot unit is less than the design value of the material strength used, the deflection of the mid-span unit of the tie beam is less than 1 / 600 of the span of the tie beam, and the internal force of each hanger is less than 0.4 times the ultimate tensile strength of the material used. Therefore, the following optimization constraints can be obtained.

[0126]

[0127] Among them, T i is the internal force of the key component, σ s (T) is the stress of the key component, and W(T) is the deformation response value of the key structure (i.e., the response data). The "key structure" is the tie beam, the arch rib unit, the arch foot unit, and the mid-span unit of the tie beam. T limit is the constraint condition of the internal force of the key component, f limit is the constraint condition of the stress of the key component, and w limit is the constraint condition of the deformation of the key component. "σ s (T) ≤ f d1 " means that the stress of the arch foot unit is less than the design value of the material strength f d1 , "W(T) ≤ l / 600" means that the deflection of the mid-span unit of the tie beam is less than 1 / 600 of the span of the tie beam, means that the internal force of each hanger is less than 0.4 times the ultimate tensile strength of the material used. A is the cross-sectional area of the hanger, and f d2 is the design strength of the hanger material.

[0128] Based on the above "s.t.", i.e., the optimization constraint limit, it can be obtained that the stress of the arch springing unit should be less than 270 MPa, the deflection of the mid-span unit of the tie beam should be less than 0.25 m, and the internal force of each hanger should be less than 1699.2 kN.

[0129] At this time, taking the preset strain information of the target bridge as the optimization objective, the optimization constraint limit as the optimization constraint condition (the above "s.t."), and the candidate internal force data as the optimization variable, the fitness can be determined. For example:

[0130]

[0131] Among them, e b (T) is the bending strain energy of the tie beam, and e r (T) is the bending strain energy of the arch rib. h(T) is an index for evaluating whether the internal force distribution of the hangers is uniform, and f * is the fitness.

[0132] Optionally, e b (T), e r (T), and h(T) can be calculated according to the following formulas:

[0133]

[0134] Among them, M i (T) is the bending moment of element i, l i is the length of element i, EI is the flexural rigidity, and σ(T) is the mean square deviation of the internal forces of this group of hangers. Element i is the element such as "tie beam, arch rib" mentioned above.

[0135] The above is the method for calculating the fitness, that is, the calculated f * is the fitness. In this embodiment, for calculating any "fitness", it can be calculated according to the above method.

[0136] Optionally, the calculation of the fitness can also be directly performed by the bridge response prediction model. For example, when training the neural network, the output is directly determined as the sample response data set and the fitness corresponding to each sample response data in the sample response data set.

[0137] In this embodiment, calculating the fitness can determine the degree of adaptation between the internal force data set and the response data set corresponding to the internal force data set, that is, determine the fitness corresponding to each internal force data in the internal force data set, providing a basis for determining or selecting better internal force data for subsequent iterative updates. At the same time, the superiority and inferiority of each internal force data can also be determined based on the fitness, facilitating the comparison of different internal force data and making the superiority and inferiority of the data simply and intuitively displayed.

[0138] S209. Determine the target fitness according to all candidate fitnesses, and determine the next internal force dataset according to the candidate internal force data corresponding to the target fitness.

[0139] Specifically, after obtaining all candidate fitnesses, the target fitness can be determined according to the following steps:

[0140] (1) Obtain the fitness condition.

[0141] Among them, the fitness condition is a pre-set condition. Optionally, the candidate fitnesses can be sorted from small to large, starting from the smallest fitness, and taking the first pre-set number of the smallest fitnesses.

[0142] (2) Use the candidate fitnesses that meet the fitness condition among all the candidate fitnesses as the target fitness.

[0143] Specifically, assume that the fitness threshold is to take the first 5 smallest fitnesses. At this time, the fitness ranking is 0.5, 1.0, 1.2, 2, 2.5, 2.7, 3.0, 3.4, 4.0, 5.0, then the first 5 fitnesses can be taken as the target fitness.

[0144] Specifically, after determining the target fitness, the candidate internal force data corresponding to the target fitness can be used as the data in the next internal force dataset.

[0145] Optionally, the Euclidean distance between the candidate internal force data can also be directly calculated, and the candidate internal force data with a small fitness and a large Euclidean distance are preferentially selected as the data in the next internal force dataset.

[0146] S210. Use the next internal force dataset as the current internal force dataset; return to execute S205.

[0147] Specifically, when obtaining the next internal force dataset, it is determined that the update iteration of the dataset is completed. Therefore, the next internal force dataset can be used as the current internal force dataset, and the step of inputting the current internal force dataset and the structural information of the target bridge into the pre-trained bridge response prediction model to obtain the current response dataset is returned until the iteration termination condition is met.

[0148] In this embodiment, for all updated current internal force datasets, there is no need to use the finite element model to determine the response dataset, but directly determine the corresponding current response dataset based on the bridge response prediction model. This not only realizes the continuous update of the internal force dataset, ensures the diversity of the data combination of the hanger internal force during the optimization process, but also ensures that no large computing power and time costs are consumed. While ensuring the accurate design of the hanger internal force of the target bridge, it can also quickly determine the better or optimal hanger internal force.

[0149] Figure 4The structural schematic diagram of the device for determining the internal force of the suspension rod provided by the embodiment of the present invention. As Figure 4 shown, the device includes:

[0150] An acquisition module 401, configured to acquire an initial internal force data set corresponding to the suspension rod of the target bridge, and use the initial internal force data set as the current internal force data set.

[0151] An output module 402, configured to input the current internal force data set and the structural information of the target bridge into a pre-trained bridge response prediction model to obtain a current response data set, where the bridge response prediction model is a model trained based on a neural network and used to simulate the finite element model of the target bridge, and the finite element model of the target bridge is a simulation model determined based on the design requirements and design scheme of the target bridge.

[0152] A judgment module 403, configured to, if the iteration termination condition is satisfied, determine the current internal force data in the current internal force data set as the internal force of the suspension rod of the target bridge; if the iteration termination condition is not satisfied, determine the next internal force data set according to the current internal force data set, the current response data set, and the bridge response prediction model, and use the next internal force data set as the current internal force data set, and return to execute the step of inputting the current internal force data set and the structural information of the target bridge into the pre-trained bridge response prediction model to obtain the current response data set.

[0153] Optionally, the device further includes a model training module; for the training process of the bridge response prediction model, the model training module is specifically configured to:

[0154] Acquire a sample internal force data set, input the sample internal force data set into the finite element model to obtain an actual response data set; input the sample internal force data set and the structural information into the neural network to obtain a sample response data set; determine the training error according to the sample response data set and the actual response data set, and determine whether the training error meets the preset error; if the training error does not meet the preset error, update the model parameters of the neural network, and return to execute the step of inputting the sample internal force data set and the structural information into the neural network to obtain the sample response data set; if the training error meets the preset error, determine the neural network as the bridge response prediction model.

[0155] Optionally, the acquisition module 401 is specifically configured to:

[0156] Determine the internal force constraint condition of the suspension rod according to the finite element model and the preset design parameters; randomly generate an initial internal force data set according to the internal force constraint condition of the suspension rod.

[0157] Optionally, to determine the next internal force data set according to the current internal force data set, the current response data set, and the bridge response prediction model, the judgment module 403 is specifically configured to:

[0158] Determine candidate internal force data and their corresponding candidate response data according to the current internal force data set, the current response data set, and the bridge response prediction model; for each candidate internal force data and its corresponding candidate response data, determine the candidate fitness corresponding to each candidate internal force data according to the candidate internal force data and its corresponding candidate response data; determine the target fitness according to all candidate fitnesses, and determine the next internal force data set according to the candidate internal force data corresponding to the target fitness.

[0159] Optionally, when determining candidate internal force data and their corresponding candidate response data according to the current internal force data set, the current response data set, and the bridge response prediction model, the determination module 403 is specifically configured to:

[0160] Obtain the preset initialization control parameters; determine the intermediate internal force data set according to the initialization control parameters, the hanger internal force constraint conditions, and the current internal force data set; input the intermediate internal force data set and the structural information into the bridge response prediction model to obtain the intermediate response data set; determine each data in the current internal force data set and the intermediate internal force data set as candidate internal force data; and determine each data in the current response data set and the intermediate response data set as candidate response data.

[0161] Optionally, when determining the candidate fitness corresponding to each candidate internal force data according to the candidate internal force data and their corresponding candidate response data, the determination module 403 is specifically configured to:

[0162] Determine the optimization constraint limit according to the hanger internal force constraint conditions and the candidate response data; determine the candidate fitness with the preset strain information of the target bridge as the optimization objective, the optimization constraint limit as the optimization constraint condition, and the candidate internal force data as the optimization variable.

[0163] Optionally, when determining the target fitness according to all candidate fitnesses, the determination module 403 is specifically configured to:

[0164] Obtain the fitness condition; use the candidate fitnesses that meet the fitness condition among all candidate fitnesses as the target fitness.

[0165] The device for determining the hanger internal force provided by the embodiments of the present invention can execute the method for determining the hanger internal force provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0166] Figure 5Schematic diagram of the electronic device 10 provided by an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described herein and / or claimed.

[0167] As Figure 5 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. The memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other via a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0168] Multiple components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disc, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0169] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the method for determining the internal force of the boom.

[0170] In some embodiments, the method for determining the internal force of the suspension rod can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for determining the internal force of the suspension rod described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the method for determining the internal force of the suspension rod by any other suitable means (e.g., by means of firmware).

[0171] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuitry, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0172] The computer programs for implementing the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer programs are executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer programs can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0173] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0174] In order to provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0175] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0176] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The relationship between the client and the server is created by computer programs that run on respective computers and have a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, and solves the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0177] An embodiment of the present invention also provides a computer program product, including a computer program which, when executed by a processor, implements the method for determining the internal force of the boom as provided in any embodiment of the present invention.

[0178] In the process of implementing the computer program product, the computer program code for performing the operations of the present invention can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0179] It should be understood that various forms of the flow shown above can be used, with steps reordered, added, or deleted. For example, the steps described in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0180] The above specific embodiments do not constitute a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for determining the internal force of a suspension rod, characterized in that The method includes: Obtaining an initial internal force data set corresponding to the suspenders of the target bridge, and using the initial internal force data set as the current internal force data set; Inputting the current internal force data set and the structural information of the target bridge into a pre-trained bridge response prediction model to obtain a current response data set, where the bridge response prediction model is a model trained based on a neural network and used to simulate the finite element model of the target bridge, and the finite element model of the target bridge is a simulation model determined based on the design requirements and design scheme of the target bridge; If the iteration termination condition is satisfied, determining the current internal force data in the current internal force data set as the internal force of the suspenders of the target bridge; If the iteration termination condition is not satisfied, determining a next internal force data set according to the current internal force data set, the current response data set and the bridge response prediction model, and using the next internal force data set as the current internal force data set, and returning to execute the step of inputting the current internal force data set and the structural information of the target bridge into the pre-trained bridge response prediction model to obtain a current response data set.

2. The method for determining the internal force of the suspension rod according to claim 1, wherein The training process of the bridge response prediction model includes: Obtaining a sample internal force data set, and inputting the sample internal force data set into the finite element model to obtain an actual response data set; Inputting the sample internal force data set and the structural information into the neural network to obtain a sample response data set; Determining a training error according to the sample response data set and the actual response data set, and determining whether the training error meets a preset error; If the training error does not meet the preset error, updating the model parameters of the neural network, and returning to execute the step of inputting the sample internal force data set and the structural information into the neural network to obtain a sample response data set; If the training error meets the preset error, determining the neural network as the bridge response prediction model.

3. The method for determining the internal force of the suspension rod according to claim 2, characterized in that, The obtaining of the initial internal force data set corresponding to the suspenders of the target bridge includes: Determining the internal force constraint conditions of the suspenders according to the finite element model and preset design parameters; Randomly generating the initial internal force data set according to the internal force constraint conditions of the suspenders.

4. The method for determining the internal force of the suspension rod according to claim 1, wherein The determining of the next internal force data set according to the current internal force data set, the current response data set and the bridge response prediction model includes: Determining candidate internal force data and their corresponding candidate response data according to the current internal force data set, the current response data set and the bridge response prediction model; For each candidate internal force data and its corresponding candidate response data, determining the candidate fitness corresponding to each candidate internal force data according to the candidate internal force data and its corresponding candidate response data; Determining a target fitness according to all the candidate fitnesses, and determining the next internal force data set according to the candidate internal force data corresponding to the target fitness.

5. The method for determining the internal force of the suspension rod according to claim 4, characterized in that The determining of the candidate internal force data and their corresponding candidate response data according to the current internal force data set, the current response data set and the bridge response prediction model includes: Obtaining a preset initialization control parameter; Determine an intermediate internal force dataset according to the initialization control parameters, the internal force constraint conditions of the suspenders, and the current internal force dataset; Input the intermediate internal force dataset and the structural information into the bridge response prediction model to obtain an intermediate response dataset; Determine each data in the current internal force dataset and the intermediate internal force dataset as the candidate internal force data; and determine each data in the current response dataset and the intermediate response dataset as the candidate response data.

6. The method for determining the internal force of the suspension rod according to claim 5, characterized in that, The determining of the candidate fitness corresponding to each candidate internal force data according to the candidate internal force data and its corresponding candidate response data includes: Determine the optimization constraint limit according to the internal force constraint conditions of the suspenders and the candidate response data; Taking the preset strain information of the target bridge as the optimization objective, the optimization constraint limit as the optimization constraint condition, and the candidate internal force data as the optimization variable, determine the candidate fitness.

7. The method for determining the internal force of the suspension rod according to claim 4, characterized in that, The determining of the target fitness according to all the candidate fitnesses includes: Obtain the fitness condition; Take the candidate fitnesses that meet the fitness condition among all the candidate fitnesses as the target fitness.

8. The method for determining the internal force of the suspension rod according to claim 1, characterized in that The iteration termination condition includes at least one of the following: The fitness corresponding to the data in the current internal force dataset meets the preset fitness requirement; The number of iterations of the current internal force dataset exceeds the preset number of times.

9. An electronic device, characterized in that, including: One or more processors; A memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method for determining the internal force of the suspenders according to any one of claims 1 to 8.

10. A readable storage medium, on which a computer program is stored, characterized in that, When the program is executed by the processor, it implements the method for determining the internal force of the suspenders according to any one of claims 1 to 8.

Images