A large special-shaped steel structure support unloading analysis method based on data processing

Abstract

The present invention discloses a method for analyzing the support unloading of large special-shaped steel structures based on data processing, belonging to the field of support unloading. The method includes constructing a connection topology diagram of the special-shaped steel structure with component connection points as nodes, and based on the division structure during the installation of the special-shaped steel structure, dividing the connection topology diagram into several unloading areas to obtain a connection topology diagram marked with unloading area information; marking the positions of temporary supports in the connection topology diagram marked with unloading area information, marked as support points, to obtain a support unloading topology diagram; according to the support unloading topology diagram, taking the unloading area as a unit, respectively constructing the mapping relationship between the change of the support force of each support point and the deformation of the nodes except the support points; according to the mapping relationship between the change of the support force of each support point and the deformation of the nodes except the support points, using a deep reinforcement learning network model to solve the support unloading strategy. The present invention solves the problem that the existing unloading methods cannot balance the deformations of each node and the total deformation.

Description

Technical Field

[0001] The present invention belongs to the field of support unloading, and particularly relates to a method for analyzing the support unloading of large special-shaped steel structures based on data processing. Background Art

[0002] During the construction process of steel structures, the support system is a key temporary facility to ensure the stability and safety of the structure during the construction stage. After the main body of the steel structure is constructed, it is necessary to carry out support unloading operations to gradually transform the stress state of the structure from relying on temporary supports to relying on its own structural system to bear all loads.

[0003] Traditional steel structure support unloading methods often rely on the empirical judgment of construction workers and lack precise mechanical analysis and real-time monitoring means. In past small-scale and simple-structured steel structure projects, experienced construction workers could quickly make judgments and implement unloading operations based on their accumulated experience without the need for complex equipment and preliminary preparations, and the cost was relatively low. For example, in the construction of small steel structures for rural self-built houses, the construction team can directly carry out support unloading based on their experience in multiple similar projects, and the operation process is simple. However, it highly depends on the personal ability of construction workers and lacks scientific and accurate analysis. There are differences in the empirical judgments of different construction workers, and it is difficult to accurately grasp the mechanical changes of the structure during the unloading process when facing large and complex structures. Once the judgment is incorrect, it is extremely easy to cause local instability of the structure. For example, in the unloading of the steel structure of some small stadiums with unique shapes, local bending deformation of the steel beam occurred due to incorrect empirical judgment. At the same time, this method cannot predict potential risks of the structure in advance and cannot form a systematic unloading plan.

[0004] Or through finite element analysis software, the whole unloading process can be simulated before construction to know in advance the deformation and stress distribution of the structure at different unloading stages, and accordingly optimize the unloading plan to improve the safety and scientificity of unloading. For example, before the unloading of a super high-rise steel structure building, finite element simulation can accurately plan the unloading sequence and quantity. At the same time, with the help of real-time monitoring by high-precision sensors, abnormal conditions of the structure can be detected in a timely manner, providing accurate data for construction decisions to ensure the safety and controllability of the unloading process. In the unloading of the steel structure of large stadiums, real-time monitoring can ensure the safety of the structure under complex forces.

[0005] However, the upfront investment cost is high. It requires professional finite element analysis software and technical personnel who are proficient in software operation, and the cost of purchasing software licenses and personnel training is not low. At the same time, a large amount of funds are also required for the layout of high-precision sensors and the construction of data acquisition systems. In addition, the simulation analysis depends on the establishment of an accurate model. If the model parameters are set unreasonably, the simulation results will deviate greatly from the actual situation, affecting the reliability of the unloading plan.

[0006] During the unloading process, if the deformation and stress changes of the structure cannot be accurately grasped, it is very easy to cause local instability of the structure, damage to components, and even lead to safety accidents of the overall structure. In addition, due to the great differences in the structural forms, load distributions, and construction conditions of different steel structure projects, the general unloading methods are difficult to meet the needs of all projects. Therefore, there is an urgent need for a more scientific, accurate, and adaptable steel structure support unloading technology. Summary of the Invention

[0007] Aiming at the above deficiencies in the prior art, a large-scale special-shaped steel structure support unloading analysis method based on data processing provided by the present invention solves the problem that the existing unloading methods cannot balance the deformations of each node and the total deformation.

[0008] To achieve the above invention purpose, the technical solution adopted by the present invention is: a large-scale special-shaped steel structure support unloading analysis method based on data processing, including:

[0009] Construct a connection topology diagram of the special-shaped steel structure with the component connection points as nodes, and based on the divided structure during the installation of the special-shaped steel structure, divide the connection topology diagram into several unloading areas to obtain a connection topology diagram marked with unloading area information;

[0010] Mark the positions of the temporary supports in the connection topology diagram marked with unloading area information, and mark them as support points to obtain a support unloading topology diagram;

[0011] According to the support unloading topology diagram, taking the unloading area as a unit, respectively construct the mapping relationship between the change of the support force of each support point and the deformation of the nodes except the support points;

[0012] According to the mapping relationship between the change of the support force of each support point and the deformation of the nodes except the support points, use the deep reinforcement learning network model to solve the support unloading strategy.

[0013] Further, the step of constructing the mapping relationship between the change of the support force of each support point and the deformation of the nodes except the support points according to the support unloading topology diagram, taking the unloading area as a unit, specifically is:

[0014] According to the support unloading topology diagram, obtain the connection information between each support point and each node in each unloading area:

[0015]

[0016] Among them, is the connection information between the th support point and the th node; is the shortest path step length on the connection line between the th support point and the th node; is the The set of nodes passed by the shortest path on the connection line between a support point and the th node;

[0017] Model the special-shaped steel structure, use simulation technology to simulate the working conditions of the change of the supporting force of each support point, and obtain the responses of each node under different working conditions;

[0018] According to the connection information between each support point and each node and the responses of each node under different working conditions, obtain the mapping relationship between the change of the supporting force of each support point and the deformation of the nodes except the support points.

[0019] Further, the obtaining of the mapping relationship between the change of the supporting force of each support point and the deformation of the nodes except the support points according to the connection information between each support point and each node and the responses of each node under different working conditions is specifically as follows:

[0020] According to the responses of each node under different working conditions, obtain the deformation conditions of each node when the supporting force of a single support point decreases;

[0021] According to the connection information between each support point and each node and the deformation conditions of each node when the supporting force of a single support point decreases, obtain the mapping relationship between the change of the supporting force of each support point and the deformation of the nodes except the support points.

[0022] Further, the expression of the mapping relationship between the change of the supporting force of each support point and the deformation of the nodes except the support points is:

[0023]

[0024] Among them, is the mapping relationship between the th support point and the th node deformation; is the th support point when the supporting force of the th support point decreases, the th node passed by the shortest path from the th node; is the deformation of the th node when the supporting force of the th support point decreases.

[0025] Further, the solving of the support unloading strategy by using the deep reinforcement learning network model according to the mapping relationship between the change of the supporting force of each support point and the deformation of the nodes except the support points is specifically as follows:

[0026] According to the mapping relationship between the change of the supporting force of each support point and the deformation of the nodes except the support points, solve the combined influence of the change of the supporting force of each support point on the deformation of the nodes except the support points:

[0027]

[0028] Among them, is the combined influence of the change in the supporting force of each support point on the deformation of the th node; is the influence weakening coefficient of the th support point with a change in the supporting force; is the single-step influence coefficient of the th support point with a change in the supporting force to the th node; is the th support point with a change in the supporting force and the th node on the shortest path step length of the connecting line; is the deformation of the th node when the supporting force of the th support point decreases; is the single-step influence coefficient of the th support point to the th node; is the node number on the shortest path from the th support point to the th support point to the th node when the supporting force of the th support point decreases; is the error correction coefficient of the th support point to the th node; is the deformation of the th node when the supporting force of the th support point decreases;

[0029] According to the combined influence of the change in the supporting force of each support point on the deformation of the nodes except the support points, the special-shaped steel structure is used as an agent in the deep reinforcement learning network model, and the state space, candidate action space, and reward of the special-shaped steel structure are constructed respectively. According to the state space, candidate action space, and reward of the special-shaped steel structure, the support unloading strategy is solved.

[0030] Furthermore, the expression of the state space is:

[0031]

[0032] Among them, is the state space at time is the set of support points to be unloaded; is the total deformation of the steel structure; is the influence weight of the action at time is The final execution action at the moment; is the combined influence of the changes in the support forces of each support point at the moment on the deformation of the th node; is the support point numbers unloaded in the final execution action at the moment; is the influence weakening coefficient of the th support point with changing support force; is the single-step influence coefficient from the th support point with changing support force to the th node; is the shortest path step length on the connection line between the

[0033] Furthermore, the expression of the candidate action space is:

[0034]

[0035] where is the candidate action space at the moment, representing and all possible combinations; is the number of support points to be unloaded; is the selected from the set of support points to be unloaded support points; is the upper limit of support point unloading; is the maximum value of batch support point unloading; is the number of support points to be unloaded; is the th node's deformation; is the node deformation warning value.

[0036] Furthermore, the solution method for the maximum value of batch support point unloading is:

[0037] B1. Using the bisection method, select support points for unloading; is the upper limit of the number of support points, with the initial value being the total number of support points in the current unloading area;

[0038] B2. Model the special-shaped steel structure, use simulation technology to select different support points for unloading, and calculate the total deformation of each node under each working condition to obtain the error between the maximum total deformation and the minimum total deformation; under each working condition, the deformation of each node is less than the node deformation warning value;

[0039] B3. Determine whether the error is within the threshold. If so, unload the maximum value of as the batch support point, otherwise, update to , and return to B1.

[0040] Furthermore, the expression of the reward is:

[0041]

[0042]

[0043] where is the reward; is the normalization function; is the upper limit of the deformation deviation; is the deformation of the th node after the final execution action is performed at time is the average deformation of each node after the final execution action is performed at time is the total number of nodes at time is the node deformation warning value; is the objective function of the support unloading task; is the first objective, indicating the minimization of the deformation fluctuation of each node; is the deformation of the th node; is the average deformation of each node; is the second objective, indicating the minimization of the total deformation of the steel structure; is the deformation fluctuation weight; is the deformation warning weight; is the deformation weight.

[0044] Furthermore, solving the support unloading strategy according to the state space, candidate action space, and reward of the special-shaped steel structure is specifically as follows:

[0045] A1. Determine the current unloading area, and initialize the Q network, target network, experience pool, and state space of the deep reinforcement learning network model;

[0046] A2. Obtain the candidate action space according to the current state space;

[0047] A3. Use the Q network to calculate the action values of all candidate actions in the candidate action space;

[0048] A4. Select the final execution action according to the action values of all candidate actions using the greedy strategy;

[0049] A5. Execute the final execution action, feedback the reward, and obtain the updated state space;

[0050] A6. According to the current state space , the final execution action , the reward and the next state space , obtain the state transition sequence, and store the state transition sequence in the experience pool;

[0051] A7. Select several groups of state transition sequences from the experience pool as training samples;

[0052] A8. Update the Q network and the target network using the training samples;

[0053] A9. Return to step A2 until all the support points in the current unloading area are unloaded, and obtain the deep reinforcement learning network model corresponding to the current unloading area, that is, the support point unloading strategy acquisition model for the current unloading area;

[0054] A10. Return to step A1 until all the support points in all unloading areas are unloaded, and obtain the support point unloading strategy acquisition models for each unloading area;

[0055] A11. Use the support point unloading strategy acquisition models for each unloading area to obtain the support unloading strategies for each unloading area respectively.

[0056] The beneficial effects of the present invention are as follows: Based on the structural division during installation, through the mapping relationship between the support points and the node deformations, relying on the deep reinforcement learning network, the unloading strategies for each unloading area are solved. The method of regional unloading takes into account the overall balance of the large steel structure. Secondly, the deformation balance degree of the nodes is considered, avoiding the significant instability of a certain part of the steel structure after the support unloading, ensuring the overall stability of the steel structure, and this stability is achieved while minimizing the impact of the support unloading on the steel structure, realizing the balance between the two. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is the flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0058] The following describes the specific embodiments of the present invention to facilitate those skilled in the art to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are within the scope of protection.

[0059] For example Figure 1As shown, in an embodiment of the present invention, a method for analyzing the support unloading of large special-shaped steel structures based on data processing includes:

[0060] Construct a connection topology diagram of the special-shaped steel structure with component connection points as nodes, and based on the division structure during the installation of the special-shaped steel structure, divide the connection topology diagram into several unloading areas to obtain a connection topology diagram marked with unloading area information;

[0061] Mark the positions of the temporary supports in the connection topology diagram marked with unloading area information, denoted as support points, to obtain a support unloading topology diagram;

[0062] According to the support unloading topology diagram, taking the unloading area as a unit, respectively construct the mapping relationship between the change in the support force of each support point and the deformation of the nodes except the support points;

[0063] According to the mapping relationship between the change in the support force of each support point and the deformation of the nodes except the support points, use the deep reinforcement learning network model to solve the support unloading strategy.

[0064] In this embodiment, the large special-shaped steel structure is divided into several installation areas during installation and installed separately. During unloading, it can be unloaded in different areas based on the area division during installation.

[0065] The specific method of constructing the mapping relationship between the change in the support force of each support point and the deformation of the nodes except the support points according to the support unloading topology diagram is as follows:

[0066] According to the support unloading topology diagram, obtain the connection information between each support point and each node in each unloading area:

[0067]

[0068] Among them, is the connection information between the th support point and the th node; is the shortest path step length on the connection line between the th support point and the th node; is the set of nodes passed by the shortest path on the connection line between the th support point and the th node;

[0069] Model the special-shaped steel structure and use simulation technology to simulate the working conditions of the change in the support force of each support point to obtain the responses of each node under different working conditions;

[0070] According to the connection information between each support point and each node and the responses of each node under different working conditions, obtain the mapping relationship between the change in the support force of each support point and the deformation of the nodes except the support points.

[0071] In this embodiment, the support points are the points of temporary support, and the nodes are the points other than the support points. The shortest path from a support point to a node is the direct influence propagation path from the support point to the node.

[0072] The mapping relationship between the change in the support force of each support point and the deformation of the nodes other than the support points is obtained according to the connection information between each support point and each node and the response of each node under different working conditions, specifically as follows:

[0073] According to the response of each node under different working conditions, obtain the deformation of each node when the support force of a single support point decreases;

[0074] According to the connection information between each support point and each node and the deformation of each node when the support force of a single support point decreases, obtain the mapping relationship between the change in the support force of each support point and the deformation of the nodes other than the support points.

[0075] In this embodiment, the mapping relationship between the change in the support force of each support point and the deformation of the nodes other than the support points describes the deformation of each node in the shortest path between the support point and the node after a single support point is unloaded. Based on this, the fitting formula of the deformation influence propagated from the support point to the node can be obtained.

[0076] The expression of the mapping relationship between the change in the support force of each support point and the deformation of the nodes other than the support points is:

[0077]

[0078] Among them, is the mapping relationship between the deformation of the th support point and the th node; is the rd node passed by the shortest path from the th support point to the th node when the support force of the th support point decreases; is the deformation of the th node when the support force of the th support point decreases.

[0079] The support unloading strategy is solved by using a deep reinforcement learning network model according to the mapping relationship between the change in the support force of each support point and the deformation of the nodes other than the support points, specifically as follows:

[0080] According to the mapping relationship between the change in the support force of each support point and the deformation of the nodes other than the support points, solve the joint influence of the change in the support force of each support point on the deformation of the nodes other than the support points:

[0081]

[0082] Among them, is the combined influence of the changes in the supporting forces of each support point on the deformation of the th node; is the influence weakening coefficient of the th support point with a change in the supporting force; is the single-step influence coefficient of the th support point with a change in the supporting force to the th node; is the shortest path step length on the connection line between the th support point with a change in the supporting force and the th node; is the deformation of the th node when the supporting force of the th support point decreases; is the single-step influence coefficient of the th support point to the th node; is the node number on the shortest path from the th support point to the th node when the supporting force of the th support point decreases; is the error correction coefficient of the th support point to the th node; is the deformation of the th node when the supporting force of the th support point decreases;

[0083] According to the combined influence of the changes in the supporting forces of each support point on the deformation of the nodes other than the support points, the special-shaped steel structure is used as an agent in the deep reinforcement learning network model, and the state space, candidate action space, and reward of the special-shaped steel structure are constructed respectively. Then, according to the state space, candidate action space, and reward of the special-shaped steel structure, the support unloading strategy is solved.

[0084] In this embodiment, the fitting formula for the deformation influence propagated from the support point to the node can obtain . Since during the support unloading process, after the temporary support is unloaded, the support point will become an ordinary node. Based on the fitting formula for the deformation influence propagated from the support point to the node, the deformation at the support point that has become a node can be given. When is 0, it means that the unloading of the

[0085] The expression of the state space is as follows:

[0086]

[0087] where, is the state space at time is the set of support points to be unloaded; is the total deformation of the steel structure; is the influence weight of the action at time ; is the final executed action at time is the combined influence of the change in the support force of each support point at time on the deformation of the is the number of the support point unloaded in the final executed action at time is the th weakening coefficient of the influence of the support point with the th change in support force; is the single-step influence coefficient of the th support point with the change in support force to the th node; is the th shortest path step length on the connection line between the

[0088] In this embodiment, there are two requirements for the support unloading strategy: ① Unload all temporary supports; ② Ensure the unloading effect.

[0089] The expression of the candidate action space is as follows:

[0090]

[0091] where, is the candidate action space at time , representing and all possible combinations; is the number of support points to be unloaded; is the selected from the set of support points to be unloaded support points; is the upper limit of support point unloading; is the maximum value of batch support point unloading; is the number of support points to be unloaded; is the th deformation of the node; is the node deformation warning value.

[0092] In this embodiment, due to the combined influence between the support points of the steel structure, there will be a situation where the effect is best when unloading several support points simultaneously. Therefore, the candidate action space is set as the combination between the support points.

[0093] The method for solving the maximum value of the batch support point unloading is as follows:

[0094] B1. Using the bisection method, select support points for unloading; is the upper limit of the number of support points, and the initial value is the total number of support points in the current unloading area;

[0095] B2. Model the special-shaped steel structure, and use simulation technology to select different support points for unloading, and calculate the total deformation of each node under each working condition to obtain the error between the maximum total deformation and the minimum total deformation; under each working condition, the deformation of each node is less than the node deformation warning value;

[0096] B3. Judge whether the error is within the threshold. If so, take as the maximum value of the batch support point unloading. Otherwise, update to , and return to B1.

[0097] In this embodiment, taking the error between the maximum total deformation and the minimum total deformation as the judgment condition for the maximum value of the batch support point unloading is to ensure the balance when selecting the unloading strategy and avoid errors caused by extreme or special situations.

[0098] In this embodiment, the maximum value of the batch support point unloading depends strongly on the threshold. Under the current threshold, if a suitable value cannot be solved, then the threshold needs to be adjusted.

[0099] The expression of the reward is:

[0100]

[0101]

[0102] Among them, is the reward; is the normalization function; is the upper limit of the deformation deviation; is the deformation of the th node after executing the final execution action at time ; is the average deformation of each node after executing the final execution action at time ; The total number of nodes at a moment; is the node deformation warning value; is the objective function for the support unloading task; is the first objective, indicating the minimization of the deformation fluctuation of each node; is the th node's deformation; is the average deformation of each node; is the second objective, indicating the minimization of the total deformation of the steel structure; is the deformation fluctuation weight; is the deformation warning weight; is the deformation weight.

[0103] In this embodiment, a reward is constructed based on the objective function of the unloading task. The three items in the reward are respectively used to balance the distribution uniformity of the deformations of each node after the support point is unloaded, the convergence of the deformations of each node to 0 after the support point is unloaded, and the control of the total deformation after the support point is unloaded. The first objective ensures that there are no particularly unstable nodes and the deformations of all nodes are uniform. When all nodes change uniformly, the overall stability of the steel structure can be guaranteed to a certain extent; the second objective ensures that the impact of support unloading on the steel structure is minimized. The balance between the first objective and the second objective can make the final unloading strategy optimal.

[0104] The solution of the support unloading strategy according to the state space, candidate action space and reward of the special-shaped steel structure is specifically as follows:

[0105] A1. Determine the current unloading area, and initialize the Q network, target network, experience pool and state space of the deep reinforcement learning network model;

[0106] A2. Obtain the candidate action space according to the current state space;

[0107] A3. Use the Q network to calculate the action values of all candidate actions in the candidate action space;

[0108] A4. According to the action values of all candidate actions, use the greedy strategy to select the final execution action;

[0109] A5. Execute the final execution action and feedback the reward to obtain the updated state space;

[0110] A6. According to the current state space , the final execution action , the reward and the next state space , obtain the state transition sequence and store the state transition sequence in the experience pool;

[0111] A7. Select several groups of state transition sequences from the experience pool as training samples;

[0112] A8. Update the Q-network and the target network using the training samples;

[0113] A9. Return to step A2 until all the support points in the current unloading area are unloaded, and obtain the deep reinforcement learning network model corresponding to the current unloading area, that is, the model for obtaining the support point unloading strategy of the current unloading area;

[0114] A10. Return to step A1 until all the support points in all unloading areas are unloaded, and obtain the models for obtaining the support point unloading strategies of each unloading area;

[0115] A11. Use the models for obtaining the support point unloading strategies of each unloading area to obtain the support unloading strategies of each unloading area respectively.

[0116] In this embodiment, the entire process of obtaining the unloading strategy is carried out in units of unloading areas.

Claims

1. A large-scale special-shaped steel structure support unloading analysis method based on data processing, characterized in that: include: A connection topology diagram of the special-shaped steel structure is constructed with the component connection points as nodes, and based on the division structure when the special-shaped steel structure is installed, the connection topology diagram is divided into several unloading areas to obtain a connection topology diagram marked with unloading area information; Mark the position of the temporary support in the connection topology diagram marked with the unloading area information, mark it as a support point, and obtain a support unloading topology diagram; According to the support unloading topology diagram, the mapping relationship between the change of the support force of each support point and the deformation of the nodes other than the support point is constructed respectively with the unloading area as the unit; According to the mapping relationship between the change of the support force of each support point and the deformation of the nodes other than the support point, the support unloading strategy is solved by using the deep reinforcement learning network model; according to the mapping relationship between the change of the support force of each support point and the deformation of the nodes other than the support point, the support unloading strategy is solved by using the deep reinforcement learning network model, specifically: According to the mapping relationship between the change of the support force of each support point and the deformation of the nodes other than the support point, the joint influence of the change of the support force of each support point on the deformation of the nodes other than the support point is solved: in, The change of the supporting force of each supporting point The combined influence of the deformation of each node; For the The influence reduction coefficient of the supporting point with changing supporting force; For the The support point of the support force change to the The single-step influence coefficient of each node; For the The supporting point of the first supporting force change is The shortest path step length on the node connection line; For the When the support force of the first support point decreases, The deformation of each node; For the Support point to The single-step influence coefficient of each node; For the When the support force of the first support point decreases, Support point to The node number on the shortest path of nodes; For the Support point to The error correction coefficient of each node; For the When the support force of the first support point decreases, The deformation of each node; According to the joint influence of the change of supporting force of each supporting point on the deformation of nodes other than the supporting point, the special-shaped steel structure is used as the intelligent agent in the deep reinforcement learning network model. The state space, candidate action space and reward of the special-shaped steel structure are constructed respectively. Based on the state space, candidate action space and reward of the special-shaped steel structure, the support unloading strategy is solved.

2. According to claim 1, the large-scale special-shaped steel structure support unloading analysis method based on data processing is characterized in that: According to the support unloading topology diagram, the mapping relationship between the change of the support force of each support point and the deformation of the node other than the support point is constructed in units of the unloading area, specifically: According to the support unloading topology diagram, obtain the connection information between each support point and each node in each unloading area: in, For the The first support point and the Connection information of each node; For the The first support point and the The shortest path step length on the node connection line; For the The first support point and the The set of nodes that the shortest path on the node connection line passes through; Model the special-shaped steel structure, use simulation technology to simulate the working conditions of the supporting force changes of each supporting point, and obtain the response of each node under different working conditions; According to the connection information between each support point and each node and the response of each node under different working conditions, the mapping relationship between the change of the support force of each support point and the deformation of the nodes except the support point is obtained.

3. According to claim 2, the large-scale special-shaped steel structure support unloading analysis method based on data processing is characterized in that: According to the connection information between each support point and each node and the response of each node under different working conditions, the mapping relationship between the change of the support force of each support point and the deformation of the node other than the support point is obtained, which is specifically: According to the response of each node under different working conditions, the deformation of each node is obtained when the support force of a single support point decreases; According to the connection information between each support point and each node and the deformation of each node when the support force of a single support point decreases, a mapping relationship between the change of the support force of each support point and the deformation of the nodes other than the support point is obtained.

4. According to claim 3, the large-scale special-shaped steel structure support unloading analysis method based on data processing is characterized in that: The expression for the mapping relationship between the change of the supporting force of each supporting point and the deformation of the nodes other than the supporting point is: in, For the The first support point and the The mapping relationship of the deformation of each node; For the When the support force of the first support point decreases, Support point to The shortest path through the nodes nodes; For the When the support force of the first support point decreases, The deformation of a node.

5. According to claim 1, the large-scale special-shaped steel structure support unloading analysis method based on data processing is characterized in that: The expression of the state space is: in, for The state space at the moment; is a set of support points to be unloaded; is the total deformation of the steel structure; for Action at all times The weight of influence; for The final execution action at the moment; for The change of the support force of each support point at the moment The combined influence of the deformation of each node; for The number of the support point unloaded in the final execution action at the moment; For the The influence reduction coefficient of the supporting point with changing supporting force; For the The support point of the support force change to the The single-step influence coefficient of each node; For the The supporting point of the first supporting force change is The shortest path step length on the node connection line.

6. The large-scale special-shaped steel structure support unloading analysis method based on data processing according to claim 1 is characterized in that: The expression of the candidate action space is: in, for The candidate action space at time , denoted and All possible combinations; is the number of support points for unloading; To be unloaded from the support point collection Selected Support points; Unload the upper limit for the support point; Unload the maximum value for the batch support point; is the number of support points to be unloaded; For the The deformation of each node; It is the node deformation warning value.

7. The large-scale special-shaped steel structure support unloading analysis method based on data processing according to claim 6 is characterized in that: The solution method for the maximum value of batch support point unloading is: B1. Use dichotomy to select Support points for unloading; The upper limit of the number of support points, the initial value is the total number of support points in the current unloading area; B2. Model the special-shaped steel structure and use simulation technology to select different The support points are unloaded, and the total deformation of each node under each working condition is calculated to obtain the error between the maximum total deformation and the minimum total deformation; under each working condition, the deformation of each node is less than the node deformation warning value; B3. Determine whether the error is within the threshold. If so, As the batch support point unload the maximum value, otherwise, Updated to , and returns B1.

8. The large-scale special-shaped steel structure support unloading analysis method based on data processing according to claim 1 is characterized in that: The reward expression is: in, For reward; is the normalization function; is the upper limit of deformation deviation; for After the final execution action is executed, The deformation of each node; for The mean deformation of each node after the final execution action is executed at the moment; for The total number of nodes at the moment; is the node deformation warning value; To support the objective function of the offloading task; is the first objective, which means minimizing the deformation fluctuation of each node; For the The deformation of each node; is the mean deformation of each node; is the second objective, which means minimizing the total deformation of the steel structure; is the deformation fluctuation weight; is the deformation warning weight; is the deformation weight.

9. The large-scale special-shaped steel structure support unloading analysis method based on data processing according to claim 1 is characterized in that: The support unloading strategy is solved according to the state space, candidate action space and reward of the special-shaped steel structure, specifically: A1. Determine the current unloading area and initialize the Q network, target network, experience pool and state space of the deep reinforcement learning network model; A2. Obtain candidate action space based on the current state space; A3. Use the Q network to calculate the action values ​​of all candidate actions in the candidate action space; A4. According to the action values ​​of all candidate actions, use the greedy strategy to select the final execution action; A5. Execute the final execution action and feedback the reward to obtain the updated state space; A6. According to the current state space , final execution action ,award and the next state space , obtain the state transition sequence, and store the state transition sequence in the experience pool; A7. Select several groups of state transition sequences from the experience pool as training samples; A8. Update the Q network and target network using training samples; A9, return to step A2, until all the support points in the current unloading area are unloaded, and obtain the deep reinforcement learning network model corresponding to the current unloading area, that is, the support point unloading strategy obtaining model of the current unloading area; A10, return to step A1, until all the support points of the unloading area are unloaded, and obtain the unloading strategy model of the support points of each unloading area; A11. Use the support point unloading strategy of each unloading area to obtain the model, and obtain the support unloading strategy of each unloading area respectively.

Images