Energy-saving high-strength anti-seismic steel grid structure optimization method based on multi-disaster adaptability

Abstract

The invention provides an energy-saving high-strength anti-seismic steel grid structure optimization method based on multi-disaster adaptability, and relates to the technical field of steel grid structure optimization, and the method comprises the steps: determining K regional disaster types; searching and determining the disaster intensity of K regions; according to the K regional disaster types, the K regional disaster intensities and the initial structural form information, carrying out stress analysis to obtain a K type stress point set and a K type stress point stress value set; carrying out anomaly identification on the K types of stress point sets and the K types of stress point stress value sets; and optimizing the initial structural form information to obtain optimized structural form information. By means of the method and device, the technical problem that in the prior art, due to the fact that the influence of multi-disaster superposition is ignored in the design of the steel truss, the adaptability of the steel truss is insufficient can be solved, and the anti-disaster performance and the overall adaptability of the steel truss are improved by optimizing the structural form of the steel truss under the multi-disaster scene.

Description

Technical Field

[0001] The present application relates to the technical field of steel grid structure optimization, and in particular to an optimization method for energy-saving, high-strength, earthquake-resistant steel grid structures based on multi-disaster adaptability. Background Art

[0002] Steel grid structure is a load-bearing structure widely used in large buildings and industrial facilities. It is especially suitable for places with large spans and high load-bearing requirements. It is composed of a series of steel pipes or steel materials, and is connected by nodes to form a grid structure with excellent mechanical properties and high seismic resistance. In the same area, multiple disasters (such as earthquakes, wind disasters, floods, etc.) may occur and affect each other. However, the existing steel grid design is usually targeted at a single disaster type (such as earthquakes), ignoring the impact of multiple disasters. As a result, the structure may perform well under certain disasters, but have significant weaknesses under other disasters. In terms of dynamic design adjustment, traditional steel grid design often lacks flexibility and cannot be adjusted in real time according to regional historical disaster records and future disaster predictions. The design is usually fixed without continuous optimization based on disaster characteristics.

[0003] In summary, the existing technology has a technical problem that the steel grid design ignores the impact of multiple disasters. It may perform well under certain disasters, but has significant weaknesses under other disasters, resulting in insufficient adaptability of the steel grid. Summary of the invention

[0004] The purpose of this application is to provide an energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability, in order to solve the technical problem in the prior art that the steel grid design ignores the impact of multiple disasters, may perform well under certain disasters, but has significant weaknesses under other disasters, resulting in insufficient adaptability of the steel grid.

[0005] In view of the above problems, the present application provides an energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability, wherein the energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability includes: obtaining the initial structural morphology information of the target steel grid, the initial stress point set, the initial stress point stress tolerance threshold set and the target application area, matching the disaster type related to the target application area, and obtaining K regional disaster types, wherein K is a positive integer; using the K regional disaster types as indexes, checking the historical regional disaster record log set of the target application area; The invention discloses a method for carrying out a search analysis to determine the disaster intensity of K regions; a stress analysis is carried out according to the disaster types of K regions and the disaster intensity of K regions, as well as the initial structural morphology information, to obtain a set of K types of stress points and a set of stress values ​​of K types of stress points; based on the initial set of stress points and the set of stress tolerance threshold values ​​of the initial stress points, anomalies are identified on the set of K types of stress points and the set of stress values ​​of K types of stress points, to obtain a set of stress points to be optimized and a set of enhanced stress values ​​of stress points to be optimized, and the initial structural morphology information is optimized to obtain the optimized structural morphology information.

[0006] The technical solution provided in this application has at least the following technical effects or advantages: By acquiring the initial structural morphology information, the initial stress point set, the initial stress point stress tolerance threshold set and the target application area of ​​the target steel grid, matching the disaster type related to the target application area, and obtaining K regional disaster types, wherein K is a positive integer; using the K regional disaster types as indexes, searching and analyzing the historical regional disaster record log set of the target application area, and determining the K regional disaster intensities; performing stress analysis according to the K regional disaster types and the K regional disaster intensities, as well as the initial structural morphology information, and obtaining K types of stress point sets and K types of stress point stress value sets; based on the initial stress point set and the initial stress point stress tolerance threshold set, performing abnormal identification on the K types of stress point sets and the K types of stress point stress value sets, obtaining the stress point set to be optimized and the enhanced stress value set of the stress points to be optimized, and optimizing the initial structural morphology information, and obtaining the optimized structural morphology information. In other words, by dynamically analyzing various disaster types and intensities based on regional historical disaster records, accurate disaster adaptability design guidance is provided. By combining initial structural morphology information and node force analysis, intelligent optimization of stress points is achieved to ensure the stability of the structure in multi-disaster scenarios. The tolerance threshold of the initial stress points is utilized to accurately locate the points to be optimized, avoid over-design, and achieve optimization of the steel grid structure morphology in multi-disaster scenarios, thereby improving seismic performance and overall adaptability.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0009] Figure 1 This is a flow chart of the energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-hazard adaptability for this application.

[0010] Figure 2 This is a schematic diagram of the process of determining the disaster intensity of K regions in the energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-hazard adaptability in this application. DETAILED DESCRIPTION

[0011] This application provides an energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability, which solves the technical problem in the prior art that the steel grid design ignores the impact of multiple disasters, may perform well under certain disasters, but has significant weaknesses under other disasters, resulting in insufficient adaptability of the steel grid. By dynamically analyzing various disaster types and intensities based on regional historical disaster records, accurate disaster adaptability design guidance is provided, and the initial structural morphology information and node force analysis are combined to achieve intelligent optimization of the stress points to ensure the stability of the structure in multi-disaster scenarios. The tolerance threshold of the initial stress points is used to accurately locate the points to be optimized, avoid over-design, and achieve optimization of the steel grid structure morphology in multi-disaster scenarios, thereby improving seismic performance and overall adaptability.

[0012] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0013] For examples, please see the attached Figure 1 The present application provides an energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability, wherein the energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability specifically includes the following steps: S100: Acquire the initial structural morphology information, the initial stress point set, the initial stress point stress tolerance threshold set and the target application area of ​​the target steel grid, match the disaster type related to the target application area, and obtain K regional disaster types, where K is a positive integer.

[0014] Specifically, the initial structural design of the target steel grid is obtained from drawings and CAD models through structural analysis tools (such as SAP2000, ETABS, etc.), including initial structural morphology information, initial stress point set, initial stress point stress tolerance threshold set and target application area. The initial structural morphology information is the basic geometric shape and structural configuration of the steel grid at the initial design stage, including the size of the grid, node location, material selection, etc. The initial stress point set refers to the key point set that bears the main external force in the steel grid structure. These points are usually areas in the structure that are subjected to concentrated loads or large deformations, such as nodes and support points. The initial stress point stress tolerance threshold set refers to the maximum allowable stress value of each stress point. Exceeding this value may cause the structure to become unstable or damaged. Through structural analysis software, static and dynamic analysis of the initial structure is performed to obtain the stress conditions of each stress point of the steel grid (such as axial force and shear force at the node). According to structural design specifications, such as the "Building Structure Load Code", the stress tolerance threshold of each stress point is determined. These thresholds are calculated based on material strength and safety factor.

[0015] The target application area refers to the geographical area where the steel grid structure is actually applied. This area may be affected by various natural disasters, such as earthquakes, wind disasters, floods, etc. According to the application scenario of the target steel grid, its geographical area is determined, and the types of disasters that the area may experience are matched through the regional disaster database (such as meteorological disaster database, earthquake disaster record database, etc.). For example, if the target application area is located at the junction of the earthquake belt and the wind belt, then the possible K disaster types are earthquakes and wind disasters. The K regional disaster types represent the various types of disasters that the target application area may face. K is a positive integer representing the number of different disaster types. By accurately obtaining the initial structural morphology information and disaster types, the structural force analysis and optimization design can be carried out more effectively, thereby improving the adaptability and safety of the structure under various disaster conditions.

[0016] S200: Using K regional disaster types as indexes, searching and analyzing a set of historical regional disaster record logs of a target application area to determine K regional disaster intensities.

[0017] Specifically, K regional disaster type intensity indicators of K regional disaster types are obtained, that is, the intensity indicators corresponding to each disaster type (such as earthquakes, wind disasters, floods, etc.), which are used to quantify the severity or influence of the disaster. For example, for earthquakes, the magnitude can be used; for wind disasters, wind speed or wind pressure may be used. Using the K regional disaster type intensity indicators as indexes, the historical regional disaster record log set of the target application area is retrieved to obtain a set of K historical regional disaster type intensity indicators, that is, the intensity indicator data of each disaster type. Through the fuzzy logic method, the K historical regional disaster type intensity indicator sets are screened and filtered to reduce the interference of single data fluctuations and obtain a more stable and accurate disaster intensity estimation. Through fuzzy screening, the influence of outliers or data fluctuations on disaster intensity assessment can be eliminated. After fuzzy screening, the concentrated values ​​of K target historical regional disaster type intensity indicators are determined, that is, the most representative intensity indicator values. The intensity analysis of the concentrated screened intensity indicators is performed to obtain the final intensity value of each disaster type, that is, the K regional disaster intensity. For example, for earthquakes, the intensity can be calculated by weighted magnitude and intensity; for floods, the intensity can be calculated by weighted water level, flow and flooded area; for storms, the intensity can be calculated by wind speed and wind force level. By analyzing historical data and intensity indicators, the disaster intensity of the target area under different disaster types can be accurately assessed, providing a more comprehensive disaster risk assessment for the management of multi-hazard areas.

[0018] S300: Perform stress analysis according to K regional disaster types and K regional disaster intensities, as well as initial structural morphology information, to obtain K types of stress point sets and K types of stress point stress value sets.

[0019] Specifically, according to K regional disaster types and K regional disaster intensities, as well as the initial structural morphology information, the steel grid is subjected to stress analysis. In the historical regional disaster record log set, the regional disaster type set, the regional disaster intensity set, and the corresponding stress point set and stress point force value set are determined. The stress point represents the node in the steel grid affected by the disaster, usually represented by its coordinates or number. The force value represents the force borne by each stress point in the steel grid, usually a specific numerical value. In the data preprocessing stage, the data set is normalized or standardized so that the range of the data is suitable for the training of the neural network. The regional disaster type set and the regional disaster intensity set are used as input data, and the corresponding stress point set and stress point force value set are used as output data, which are divided into a training set and a validation set.

[0020] Construct a neural network model, including an input layer, a hidden layer, and an output layer. The input is a combination of regional disaster type and disaster intensity, such as the code of the disaster type and the numerical value of the disaster intensity; it includes multiple hidden layers to capture the complex relationship between input and output; the output is the predicted stress point and stress value, which is usually a regression problem, so the output layer uses a linear activation function. Use training data to train the model. During the training process, the neural network will continuously adjust the parameters to make the predicted results closer to the actual stress points and stress values. After the training is completed, use the test data to evaluate the performance of the model. Indicators such as the mean square error (MSE) and mean absolute error (MAE) of the model can be calculated to measure the accuracy of the prediction. After the model is trained, use the new regional disaster type and intensity data for prediction to obtain the predicted stress points and stress values.

[0021] Input K regional disaster types and K regional disaster intensities into the trained force analysis model, determine the set of force points under K disaster types, and predict the corresponding force values, which represent the magnitude of the force borne by each force point. According to the results of the force analysis, determine the set of K types of force points and the set of K types of force values ​​of force points, reflecting the specific impact of each disaster type on different force points. By comprehensively analyzing the impact of different disaster types on steel grids, the force points and force values ​​under each disaster can be accurately obtained, which helps to determine the weak links in the structure, thereby providing a basis for the structural optimization and reinforcement of steel grids, avoiding over-design or waste of resources, ensuring the adaptability of steel grids in practical applications, and enhancing their disaster resistance under extreme disasters.

[0022] S400: Based on the initial stress point set and the initial stress point force tolerance threshold set, anomalies are identified on the K types of stress point sets and the K types of stress point force value sets, to obtain the stress point set to be optimized and the enhanced force value set of the stress points to be optimized, and the initial structural morphology information is optimized to obtain the optimized structural morphology information.

[0023] Specifically, determine whether there are type stress points that do not belong to the initial stress point set in the K type stress point sets, and add them to the first stress point set to be optimized. Determine the stress values ​​of the type stress points that do not belong to the initial stress point set in the K type stress point force value sets, and add them to the first stress point enhanced stress value set to be optimized. If the K type stress point sets all belong to the initial stress point set, map and match the K type stress point sets in the initial stress point set, determine the corresponding initial stress point force tolerance threshold, calculate the difference between each type stress point set and its corresponding initial stress point force tolerance threshold set, if the difference is negative, use the difference as one of the enhanced stress values ​​of the stress point to be optimized in the second stress point enhanced force value set to be optimized, and add the type stress point to the second stress point set to be optimized. Merge the first stress point set to be optimized with the second stress point set to be optimized to obtain the stress point set to be optimized. The first enhanced force value set of the stress points to be optimized is combined with the second enhanced force value set of the stress points to be optimized to obtain the enhanced force value set of the stress points to be optimized.

[0024] Based on historical data, a structural optimizer is pre-built. The set of stress points to be optimized, the set of enhanced stress values ​​of stress points to be optimized, and the initial structural morphology information are input into the structural optimizer to obtain the optimized structural morphology information, including structural adjustment and stress point position adjustment. Verify whether the optimized structural morphology information meets the design requirements and safety standards. The optimized structural morphology information contains the optimized stress point positions, structural dimensions, and material distribution under disaster conditions, aiming to improve the disaster resistance of the overall structure. By optimizing the structural morphology and enhancing the bearing capacity of key stress points, the adaptability of steel grids in multi-disaster scenarios can be significantly improved, and damage caused by disasters can be reduced. By accurately identifying abnormal stress points and optimizing them, failure or damage of the structure in disasters can be avoided, thereby improving the safety of the overall structure.

[0025] Further, as attached Figure 2 As shown, the present application S200 includes: Obtain K regional disaster type intensity indicators of K regional disaster types; use the K regional disaster type intensity indicators as indexes to search a historical regional disaster record log set to obtain a K historical regional disaster type intensity indicator set; traverse the K historical regional disaster type intensity indicator sets to perform centralized fuzzy screening of intensity indicators to determine the K regional disaster intensities.

[0026] Specifically, the corresponding intensity index is determined for the determined K regional disaster types, that is, K regional disaster type intensity indexes. For example, the earthquake intensity index is measured by magnitude (Mw) and intensity (MMI); the wind disaster intensity index is measured by wind speed or wind force level (such as the maximum sustained wind speed of a typhoon); the snow intensity index is measured by snow depth and low temperature duration; the frost damage intensity index is measured by low temperature duration; the flood intensity index is measured by water level, flow and flooded area. The historical regional disaster record log set is a data set that records historical disaster events in the target area, including information such as the type of disaster, the time of the disaster, the intensity of the disaster, and the scope of the disaster.

[0027] According to the K regional disaster type intensity indicators, a search is performed in the historical regional disaster record log set to obtain a set of K historical regional disaster type intensity indicators, that is, a set of intensity indicators related to each disaster type in the target area. The centralized fuzzy screening method sets the fuzzy screening bandwidth and tolerates the data to obtain a relatively centralized intensity indicator. The purpose is to remove outliers or noise and retain the intensity data that meets the standards. After screening, the concentrated value of the disaster type intensity indicator of each region is obtained, and the concentrated value of the disaster type intensity indicator is analyzed to obtain the disaster intensity of each region, that is, the K regional disaster intensity. The intensity calculation method can be carried out according to the characteristics of different disaster types. For example, the intensity of an earthquake can be obtained by weighted calculation of magnitude and intensity, and the intensity of a storm can be obtained by weighted calculation of wind speed and wind force level. For example, assuming that the concentrated value of the wind speed after screening is 6.5 and the concentrated value of the wind force level is 1.3, according to the weighted formula of water level and flow velocity, assuming that the wind weight is 0.5 and the flow weight is 0.5, the flood disaster intensity is 3.9. Through the centralized fuzzy screening method, outliers and noise data can be removed, the accuracy of the final intensity index can be improved, and personalized processing can be performed according to the disaster types and intensity indicators of different regions, which can more accurately reflect the disaster risks of various regions.

[0028] Furthermore, the present application also includes the following steps: The means of the K historical regional disaster type intensity index sets are calculated respectively to obtain the means of the K historical regional disaster type intensity indexes; the fluctuation coefficients of the K historical regional disaster type intensity index sets are traversed to obtain K fluctuation coefficients, and K centralized fuzzy screening bandwidths are matched according to the sizes of the K fluctuation coefficients; the means of the K historical regional disaster type intensity indexes are used as K centralized fuzzy screening starting points, and the K historical regional disaster type intensity index sets are centralized fuzzy screened according to the K centralized fuzzy screening bandwidths to obtain K target historical regional disaster type intensity index centralized values; the K target historical regional disaster type intensity index centralized values ​​are subjected to intensity analysis to obtain K regional disaster intensities.

[0029] Specifically, for each disaster type, the mean of all its historical disaster intensity indicators is calculated. The mean reflects the average intensity of the disaster type. As the basic intensity level of the disaster type, it is usually obtained by dividing the intensity index value of each disaster type by the number of intensity indicators. According to the mean of K historical regional disaster type intensity indicators, the standard deviation of the set of K historical regional disaster type intensity indicators is calculated. By calculating the ratio of the standard deviation of the set of K historical regional disaster type intensity indicators and the mean of K historical regional disaster type intensity indicators, K fluctuation coefficients are obtained. The fluctuation coefficient is an important indicator to describe the degree of data dispersion, which is the ratio of the standard deviation of the data to the mean. The larger the fluctuation coefficient, the greater the degree of dispersion of the data, that is, the wider the range of data variation. By calculating the fluctuation coefficient, the stability of the data can be understood.

[0030] According to the size of the fluctuation coefficient, a centralized fuzzy screening bandwidth is matched for each disaster type. The larger the fluctuation coefficient, the more discrete the data, and the larger the screening bandwidth should be set to adapt to the discreteness of the data. The mean values ​​of the intensity indicators of disaster types in K historical regions are taken as the starting points for K centralized fuzzy screening, and centralized fuzzy screening is performed on the sets of intensity indicators of disaster types in K historical regions according to the K centralized fuzzy screening bandwidths. In other words, the mean value of the intensity indicator of each disaster type is taken as the starting point, and the corresponding Kaili City intensity indicator set is screened according to the matched corresponding screening bandwidth, and the concentrated value of the intensity indicator of each disaster type is determined using fuzzy logic methods. Centralized fuzzy screening refers to fuzzifying the data set, aggregating the data through a loose error range, and aggregating multiple data points into a more reasonable representative value. It is suitable for processing data with uncertainty or incomplete information.

[0031] Specifically, with K centralized fuzzy screening starting points as the center, according to the K centralized fuzzy screening bandwidths, K centralized fuzzy screening starting point neighborhoods are determined in the K historical regional disaster type intensity indicator sets. K historical regional disaster type intensity indicators are randomly selected at the edge of the K centralized fuzzy screening starting point neighborhoods as K centralized fuzzy screening iteration points, and K centralized fuzzy screening iteration point neighborhoods are determined in the K historical regional disaster type intensity indicator sets. The neighborhood density of the K centralized fuzzy screening iteration point neighborhoods and the neighborhood density of the K centralized fuzzy screening starting point neighborhoods are calculated to determine which neighborhood density is greater. If the neighborhood density of the iteration point is greater, the K historical regional disaster type intensity indicator sets are centralized fuzzy screened according to the K centralized fuzzy screening iteration points and the K centralized fuzzy screening bandwidths until the preset maximum number of screenings is met, and the K centralized fuzzy screening iteration points obtained in the last screening are used as the centralized values ​​of the K target historical regional disaster type intensity indicators.

[0032] If the neighborhood density of the starting point is greater, a random function is introduced to generate a random value, and the first random value is compared with the preset threshold. The preset random value is a fixed value (for example, 0.5), which determines the conditions for the introduction of randomness. If the first random value is greater than the preset random value, a screening path is entered; otherwise, another screening path is entered. After multiple screenings, the final concentrated value of the K historical regional disaster type intensity indicators is the precise disaster intensity value of each disaster type.

[0033] The concentrated values ​​of the intensity index of disaster types in K target historical regions are analyzed to determine the disaster intensity of each disaster type in the target region, and obtain K regional disaster intensities. For example, the concentrated values ​​of earthquake indicators are magnitude and intensity. The magnitude and intensity are weighted to obtain the regional disaster intensity. Assuming that the concentrated value of magnitude is 6.0, the concentrated value of intensity is 7.5, and the weights of magnitude and intensity are 0.6 and 0.4 respectively, the final earthquake disaster intensity is 0.6×6.0+0.4×7.5=6.3. By calculating the mean and fluctuation coefficient, the representative intensity index of each region can be effectively screened out, thereby avoiding errors caused by data fluctuations. The introduction of the fluctuation coefficient enables the model to adjust the screening bandwidth according to the data characteristics of different regions, improving adaptability and accuracy.

[0034] Furthermore, the present application also includes the following steps: According to the K centralized fuzzy screening bandwidths, K centralized fuzzy screening starting point neighborhoods of the K centralized fuzzy screening starting points are constructed in the K historical regional disaster type intensity indicator sets; K historical regional disaster type intensity indicators are randomly extracted from the edges of the K centralized fuzzy screening starting point neighborhoods as K centralized fuzzy screening iteration points, and K centralized fuzzy screening iteration point neighborhoods are constructed; it is determined whether the neighborhood density of the K centralized fuzzy screening iteration point neighborhoods is greater than or equal to the neighborhood density of the K centralized fuzzy screening starting point neighborhoods; if so, centralized fuzzy screening is performed on the K historical regional disaster type intensity indicator sets based on the K centralized fuzzy screening iteration points and the K centralized fuzzy screening bandwidths until the preset maximum screening times are met, and the K centralized fuzzy screening iteration points obtained from the last screening are used as the centralized values ​​of the K target historical regional disaster type intensity indicators.

[0035] Furthermore, the present application also includes the following steps: If not, a first random value is randomly generated based on the rand function. When the first random value is greater than the preset random value, a centralized fuzzy screening is performed on the K historical regional disaster type intensity index sets based on the K centralized fuzzy screening iteration points and the K centralized fuzzy screening bandwidths until the preset maximum number of screening times is met, and the screening result obtained from the last screening is used as the centralized value of the K target historical regional disaster type intensity indexes; when the first random value is less than or equal to the preset random value, a centralized fuzzy screening is performed on the K historical regional disaster type intensity index sets based on the K centralized fuzzy screening starting points and the K centralized fuzzy screening bandwidths until the preset maximum number of screening times is met, and the screening result obtained from the last screening is used as the centralized value of the K target historical regional disaster type intensity indexes.

[0036] Specifically, in the K historical regional disaster type intensity index sets, according to the K centralized fuzzy screening bandwidths, the K centralized fuzzy screening starting point neighborhoods of the K centralized fuzzy screening starting points are determined. For the sake of explanation, the following explanation is based on one of the intensity index sets. A strength index set is randomly selected from the K historical regional disaster type intensity index sets as the first historical regional disaster type intensity index set, the first centralized fuzzy screening starting point is determined from the K centralized fuzzy screening starting points, and the first centralized fuzzy screening bandwidth is determined from the K centralized fuzzy screening bandwidths. In the first historical regional disaster type intensity index set, with the first centralized fuzzy screening starting point as the center, a neighborhood range is determined according to the first centralized fuzzy screening bandwidth as the first centralized fuzzy screening starting point neighborhood.

[0037] From the edge of the neighborhood of the starting point of the fuzzy screening in the first set, a data point is randomly extracted as the concentrated fuzzy screening iteration point, and at the same time, the neighborhood of the first concentrated fuzzy screening iteration point of this data point is determined in the first historical regional disaster type intensity index set. The neighborhood density of the neighborhood of the starting point of the fuzzy screening in the first set and the neighborhood density of the neighborhood of the fuzzy screening iteration point in the first set are calculated. The calculation process of the neighborhood density of the neighborhood of the starting point of the fuzzy screening in the first set is as follows: count the total number of historical regional disaster type intensity indicators in the neighborhood of the starting point of the fuzzy screening in the first set, calculate the ratio of the statistical result to twice the fuzzy screening bandwidth in the first set, and obtain the neighborhood density of the neighborhood of the starting point of the fuzzy screening in the first set. The calculation process of the neighborhood density of the neighborhood of the fuzzy screening iteration point in the first set is similar.

[0038] Determine whether the density of the neighborhood of the current iteration point is greater than or equal to the neighborhood density of the initial screening point. A large neighborhood density indicates that the data in this area is tightly clustered and has a high correlation. If it is greater than, the first concentrated fuzzy screening iteration point is used as the new center point, and the concentrated fuzzy screening is continued in the first historical area disaster type intensity index set according to the first concentrated fuzzy screening bandwidth. Repeat the above process until the preset maximum number of screenings is met, and the first concentrated fuzzy screening iteration point obtained in the last screening is used as the concentrated value of the disaster type intensity index of the first target historical area. The preset maximum number of screenings refers to the preset upper limit of the number of screenings when performing centralized fuzzy screening. Usually, a maximum value is set to control the number of screening iterations to avoid infinite loops.

[0039] On the contrary, that is, the neighborhood density of the neighborhood of the fuzzy screening iteration point in the first set is less than the neighborhood density of the neighborhood of the fuzzy screening starting point in the first set. A random number between 0 and 1 is randomly generated according to the rand function as the first random value. A random value is predefined for comparison with the first random value. If the first random value is greater than the preset random value, a screening path is executed; if the first random value is less than or equal to the preset random value, another path is executed. By generating random numbers, a certain inferior solution is accepted to avoid falling into the local optimum.

[0040] The first random value is compared with the preset random value. If the first random value is greater than the preset random value, the first historical regional disaster type intensity index set is subjected to centralized fuzzy screening according to the first centralized fuzzy screening iteration point and the first centralized fuzzy screening bandwidth until the preset maximum number of screenings is met, and the screening result obtained from the last screening is used as the centralized value of the first target historical regional disaster type intensity index. Through this screening path, multiple rounds of screening are performed until the preset maximum number of screenings is met.

[0041] If the first random value is less than or equal to the preset random value, the first historical area disaster type intensity index set is concentrated and fuzzy screened according to the first centralized fuzzy screening starting point and the first centralized fuzzy screening bandwidth until the preset maximum number of screenings is met, and the screening result obtained by the last screening is used as the first target historical area disaster type intensity index centralized value. Through this screening path, multiple rounds of screening are performed until the preset maximum number of screenings is met. According to the path of the screening process, the final first target historical area disaster type intensity index centralized value is obtained.

[0042] For other historical regional disaster type intensity indicator sets of the K historical regional disaster type intensity indicator sets, the above complete steps are executed, and the mean of the K historical regional disaster type intensity indicators is used as the K centralized fuzzy screening starting points. The K historical regional disaster type intensity indicator sets are centralized fuzzy screened according to the K centralized fuzzy screening bandwidths to obtain the K target historical regional disaster type intensity indicator centralized values. After multiple screenings, the final K historical regional disaster type intensity indicator centralized values ​​are the precise disaster intensity values ​​of each disaster type. Through iterative screening and neighborhood density judgment, data can be gradually optimized in multiple screenings, noise and errors can be reduced, and the final disaster intensity assessment can be ensured to be more accurate. When the disaster data features are complex, random paths can help the system perform diversified screening under different conditions to avoid falling into local optimal solutions, thereby improving the accuracy of data processing.

[0043] Further, the present application S400 includes: Determine whether there is a type of stress point that does not belong to the initial stress point in the K type stress point sets. If so, add it to the first stress point set to be optimized, and add its corresponding type stress point stress value to the first enhanced stress value set of stress points to be optimized; if not, perform mapping matching based on the K type stress point sets and the initial stress point set. According to the mapping matching result, subtract the corresponding initial stress point force tolerance threshold in the initial stress point force tolerance threshold set from the stress value of each type stress point. When the difference is negative, use the difference as the enhanced stress value of the stress point to be optimized, and use the type stress point as the stress point to be optimized to obtain a second set of stress points to be optimized and a second set of enhanced stress values ​​of stress points to be optimized; integrate the first set of stress points to be optimized and the second set of stress points to be optimized to obtain a set of stress points to be optimized, and integrate the first set of enhanced force values ​​of stress points to be optimized and the second set of enhanced force values ​​of stress points to be optimized to obtain a set of enhanced force values ​​of stress points to be optimized.

[0044] Specifically, determine whether there are any points among the predicted K types of stress points that do not belong to the initial stress point set. If there are stress points that do not belong to the initial stress point set, they need to be optimized so that they can withstand greater disaster forces. These points will be added to the first set of stress points to be optimized, and their stress values ​​will be added to the first set of enhanced stress values ​​for stress points to be optimized. Traverse the K types of stress point sets and compare them with the initial stress point set. If a stress point does not belong to the initial stress point set, add the stress point to the first set of stress points to be optimized, and add the stress value corresponding to the stress point to the first set of enhanced stress values ​​for stress points to be optimized.

[0045] On the contrary, if all K types of stress points belong to the initial stress point set, it is necessary to further check the stress condition of each stress point. By mapping and matching each type of stress point with the initial stress point, the difference between the stress value of each type of stress point and the tolerance threshold of the initial stress point is calculated. If the difference is a negative number, it means that the stress value of the current stress point exceeds the tolerance threshold in the initial design and needs to be optimized. Map and match each type of stress point with the initial stress point to find the initial stress point corresponding to each type of stress point. For each pair of matching stress points, the difference is determined by subtracting the force tolerance threshold of the initial stress point from the stress value of the type stress point. If the difference is a negative value, the difference is used as the enhanced stress value of the stress point to be optimized, and the type of stress point is added to the second set of stress points to be optimized.

[0046] The first set of stress points to be optimized is combined with the second set of stress points to be optimized to obtain a set of stress points to be optimized. The first set of enhanced stress values ​​of stress points to be optimized is combined with the second set of enhanced stress values ​​of stress points to be optimized to obtain a set of enhanced stress values ​​of stress points to be optimized. Through accurate calculation and judgment of each stress point, those stress points that are greatly affected by disasters and need structural optimization can be accurately screened out to improve the disaster resistance of the steel grid.

[0047] Furthermore, the present application also includes the following steps: Pre-constructing a structural optimizer; using the pre-constructed structural optimizer to analyze a set of stress points to be optimized, a set of enhanced stress values ​​of stress points to be optimized, and initial structural morphology information to obtain the optimized structural morphology information.

[0048] Furthermore, the present application also includes the following steps: Acquire multiple sample stress point sets to be optimized, multiple sample stress point enhanced stress value sets to be optimized, multiple sample initial structural morphology information, and corresponding multiple sample optimized structural morphology information as training data; use the training data to supervise the training of the framework built based on the feedforward neural network, and update the network parameters during the training until the training converges, so as to obtain the trained structural optimizer.

[0049] Specifically, the sample data sets are collected and organized, including multiple sample stress point sets to be optimized, multiple sample stress point enhanced stress value sets to be optimized, multiple sample initial structural morphological information, and multiple sample optimized structural morphological information. Multiple sample stress point sets to be optimized refer to the stress point sets that need to be optimized in multiple different structural samples. These points may exceed the allowable stress range of the initial design due to the impact of disasters, and their disaster resistance needs to be improved through optimization. Multiple sample stress point enhanced stress value sets to be optimized correspond to the stress point sets to be optimized in multiple samples, indicating the stress values ​​that need to be enhanced for the stress points that need to be optimized under the influence of disasters. Multiple sample initial structural morphological information refers to the initial design morphological data of multiple steel grids or structural samples, including the geometric shape, node layout, material distribution, etc. of the structure. Multiple sample optimized structural morphological information refers to the morphological information of multiple structural samples after optimization. These structural morphological information are obtained by optimizing the stress point sets to be optimized and the enhanced stress value sets.

[0050] Preprocess the collected sample data sets, normalize or standardize the input and output data to ensure the stability and efficiency of network training, and ensure that the data format is suitable for the requirements of the input layer and output layer of the neural network. Construct a feedforward neural network, which includes an input layer, a hidden layer, and an output layer, and the data flows from the input layer to the output layer. Train the framework constructed by the feedforward neural network according to the training set. The feedforward neural network receives data from multiple samples as input and compares it with the target output (optimized structural morphology information). The set of stress points to be optimized, the set of enhanced stress values ​​of the stress points to be optimized, and the initial structural morphology information are used as inputs to the network. Through one or more layers of neurons, feature extraction and information processing are performed to learn the complex relationships in the data. Output the optimized structural morphology information. During the training process, the error between the network output and the actual optimization result (usually the mean square error or other loss function) is calculated, and the network weights and biases are updated through the back propagation algorithm until the error converges.

[0051] After multiple iterations, until the loss function of the neural network converges, it means that the network has learned how to optimize the structural morphology based on the input stress points, enhanced stress values, and initial structural morphology information. After each round of training, the error is calculated and the network parameters (weights and biases) are updated. As the training progresses, the network will gradually reduce the error and eventually converge, indicating that the model is accurate enough. Repeat the training process until the prediction performance of the network is no longer significantly improved, that is, it reaches a convergence state. After the training is completed, the neural network model is a structural optimizer, which can be used in practical applications to predict and optimize the morphology of steel grid structures based on new stress points to be optimized and enhanced stress values.

[0052] The set of stress points to be optimized, the set of enhanced stress values ​​of stress points to be optimized, and the initial structural morphology information are input into the structural optimizer. The structural optimizer is a trained neural network model. It processes the input set of stress points to be optimized, the set of enhanced stress values, and the initial structural morphology information to predict an optimized structural morphology. The set of stress points to be optimized, the set of enhanced stress values, and the initial structural morphology information are transmitted through the input layer of the network. The hidden layer in the network will extract features from the input data through weights and activation functions, and learn the relationship between structural morphology and disaster stress. The optimized structural morphology information is obtained through the output layer, including the new position of the nodes, the number of optimized nodes, the adjustment of the structural size, the optimization of material distribution, etc.

[0053] After the network has been fed-forward calculated, the output result is the optimized structural morphology information, including the adjustment of the position of the stress point, the adjustment of the structural size, the optimization of material properties, etc. For example, if a node of the initial structure is subjected to excessive force under strong winds or earthquakes, the structural optimizer will decide to increase the size of the node or change the material by analyzing the stress point to be optimized and the enhanced stress value, and finally obtain an optimized node position and material distribution. By optimizing the structural morphology, the force bearing capacity and stability of the steel grid under disaster conditions can be effectively improved, and the risk of damage in disasters can be reduced. The optimization plan can not only cope with a specific disaster (such as an earthquake), but also comprehensively consider the impact of multiple disasters to ensure the efficient operation of the steel grid structure in a complex disaster environment.

[0054] In summary, the energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability provided in this application has the following technical effects: By acquiring the initial structural morphology information, the initial stress point set, the initial stress point stress tolerance threshold set and the target application area of ​​the target steel grid, matching the disaster type related to the target application area, and obtaining K regional disaster types, wherein K is a positive integer; using the K regional disaster types as indexes, searching and analyzing the historical regional disaster record log set of the target application area, and determining the K regional disaster intensities; performing stress analysis according to the K regional disaster types and the K regional disaster intensities, as well as the initial structural morphology information, and obtaining K types of stress point sets and K types of stress point stress value sets; based on the initial stress point set and the initial stress point stress tolerance threshold set, performing abnormal identification on the K types of stress point sets and the K types of stress point stress value sets, obtaining the stress point set to be optimized and the enhanced stress value set of the stress points to be optimized, and optimizing the initial structural morphology information, and obtaining the optimized structural morphology information. In other words, by dynamically analyzing various disaster types and intensities based on regional historical disaster records, accurate disaster adaptability design guidance is provided. By combining initial structural morphology information and node force analysis, intelligent optimization of stress points is achieved to ensure the stability of the structure in multi-disaster scenarios. The tolerance threshold of the initial stress points is utilized to accurately locate the points to be optimized, avoid over-design, and achieve optimization of the steel grid structure morphology in multi-disaster scenarios, thereby improving seismic performance and overall adaptability.

[0055] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0056] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. An energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability, characterized in that: include: Obtain the initial structural morphology information, the initial stress point set, the initial stress point stress tolerance threshold set and the target application area of ​​the target steel grid, match the disaster type related to the target application area, and obtain K regional disaster types, where K is a positive integer; Using K regional disaster types as indexes, the historical regional disaster record log set of the target application area is retrieved and analyzed to determine the K regional disaster intensities; According to the K regional disaster types and K regional disaster intensities, as well as the initial structural morphology information, a stress analysis is performed to obtain a set of K types of stress points and a set of K types of stress values ​​of stress points; Based on the initial stress point set and the initial stress point force tolerance threshold set, the K types of stress point sets and the K types of stress point force value sets are identified for anomalies, the stress point set to be optimized and the enhanced stress value set of the stress points to be optimized are obtained, and the initial structural morphology information is optimized to obtain the optimized structural morphology information.

2. The energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability according to claim 1 is characterized in that: Using K regional disaster types as indexes, the historical regional disaster record log set of the target application area is retrieved and analyzed to determine the K regional disaster intensities, including: Obtain K regional disaster type intensity indicators of K regional disaster types; Using the K regional disaster type intensity indicators as indexes, searching a historical regional disaster record log set to obtain a set of K historical regional disaster type intensity indicators; The K historical regional disaster type intensity index sets are traversed to perform concentrated fuzzy screening of intensity indicators to determine the disaster intensity of the K regions.

3. The energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability according to claim 2 is characterized in that: Traversing the K historical regional disaster type intensity index sets to perform concentrated fuzzy screening of intensity indicators, and determining the K regional disaster intensities, including: Calculate the mean values ​​of the K historical regional disaster type intensity index sets respectively to obtain the mean values ​​of the K historical regional disaster type intensity indexes; Traversing and calculating the fluctuation coefficients of the K historical regional disaster type intensity index sets to obtain K fluctuation coefficients, and matching K centralized fuzzy screening bandwidths according to the sizes of the K fluctuation coefficients; Taking the mean values ​​of the K historical regional disaster type intensity indicators as K centralized fuzzy screening starting points, performing centralized fuzzy screening on the K historical regional disaster type intensity indicator sets according to the K centralized fuzzy screening bandwidths, and obtaining the centralized values ​​of the K target historical regional disaster type intensity indicators; The intensity analysis is performed on the concentrated values ​​of the disaster type intensity indicators of K target historical regions to obtain the disaster intensity of K regions.

4. The energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability according to claim 3 is characterized in that: Taking the mean values ​​of the K historical regional disaster type intensity indicators as K centralized fuzzy screening starting points, the K historical regional disaster type intensity indicator sets are centralized fuzzy screened according to the K centralized fuzzy screening bandwidths to obtain the centralized values ​​of the K target historical regional disaster type intensity indicators, including: Constructing K centralized fuzzy screening starting point neighborhoods of the K centralized fuzzy screening starting points in the K historical regional disaster type intensity index sets according to the K centralized fuzzy screening bandwidths; Randomly extracting K historical regional disaster type intensity indicators from the edges of the K centralized fuzzy screening starting point neighborhoods as K centralized fuzzy screening iteration points, and constructing K centralized fuzzy screening iteration point neighborhoods; Determine whether the neighborhood density of the K centralized fuzzy screening iteration points is greater than or equal to the neighborhood density of the K centralized fuzzy screening starting point neighborhoods. If so, perform centralized fuzzy screening on the K historical regional disaster type intensity index sets based on the K centralized fuzzy screening iteration points and the K centralized fuzzy screening bandwidths until the preset maximum number of screening times is met, and use the K centralized fuzzy screening iteration points obtained from the last screening as the centralized values ​​of the K target historical regional disaster type intensity indicators.

5. The energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability according to claim 4, characterized in that: Determining whether the neighborhood density of the K centralized fuzzy screening iteration point neighborhoods is greater than or equal to the neighborhood density of the K centralized fuzzy screening starting point neighborhoods, further comprising: If not, a first random value is randomly generated based on the rand function. When the first random value is greater than the preset random value, a centralized fuzzy screening is performed on the K historical regional disaster type intensity index sets based on the K centralized fuzzy screening iteration points and the K centralized fuzzy screening bandwidths until the preset maximum number of screening times is met, and the screening result obtained from the last screening is used as the centralized value of the K target historical regional disaster type intensity index; When the first random value is less than or equal to the preset random value, the K centralized fuzzy screening starting points and the K centralized fuzzy screening bandwidths are used to perform centralized fuzzy screening on the K historical regional disaster type intensity index sets until the preset maximum number of screening times is met, and the screening result obtained from the last screening is used as the centralized value of the K target historical regional disaster type intensity indicators.

6. The energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability according to claim 1, characterized in that: Based on the initial stress point set and the initial stress point stress tolerance threshold set, the K types of stress point sets and the K types of stress point stress value sets are identified as abnormalities to obtain a set of stress points to be optimized and a set of enhanced stress values ​​of stress points to be optimized, including: Determine whether there is a type of force point that does not belong to the initial force point in the K type force point sets. If so, add it to the first set of force points to be optimized, and add the force value of the corresponding type of force point to the first set of enhanced force values ​​of the force points to be optimized; If not, mapping matching is performed based on the K types of stress point sets and the initial stress point set. According to the mapping matching result, the stress value of each type of stress point is subtracted from the corresponding initial stress point stress tolerance threshold in the initial stress point stress tolerance threshold set. When the difference is a negative value, the difference is used as the enhanced stress value of the stress point to be optimized, and the type of stress point is used as the stress point to be optimized, to obtain the second set of stress points to be optimized and the second set of enhanced stress values ​​of the stress points to be optimized; The first set of stress points to be optimized and the second set of stress points to be optimized are integrated to obtain the set of stress points to be optimized, and the first set of enhanced force values ​​of stress points to be optimized and the second set of enhanced force values ​​of stress points to be optimized are integrated to obtain the set of enhanced force values ​​of stress points to be optimized.

7. The energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability according to claim 1, characterized in that: include: Pre-built structure optimizer; The pre-built structural optimizer is used to analyze the set of stress points to be optimized, the set of enhanced stress values ​​of the stress points to be optimized and the initial structural morphology information to obtain the optimized structural morphology information.

8. The energy-saving, high-strength, earthquake-resistant steel grid structure optimization method based on multi-disaster adaptability according to claim 7, characterized in that: include: Acquire multiple sample stress point sets to be optimized, multiple sample stress point enhanced stress value sets to be optimized, multiple sample initial structural morphology information, and corresponding multiple sample optimized structural morphology information as training data; The framework constructed based on the feedforward neural network is supervised and trained using the training data, and the network parameters are updated during the training until the training converges, thereby obtaining the trained structural optimizer.

Images