Structural parameter determination method of building structure and related equipment

Abstract

The invention provides a structure parameter determination method of a building structure and related equipment. The method comprises the following steps: acquiring a first structure parameter and a preset environment parameter about the building structure; inputting the first structure parameter into a trained response prediction model to obtain a corresponding first response; wherein the response prediction model is obtained by training based on a static analysis result and a dynamic analysis result of a building structure sample, the static analysis result comprises a static response of the building structure sample under a static load, and the dynamic analysis result comprises a vibration response of the building structure sample under a dynamic load; and updating the first structure parameter based on the first response and a preset optimization target to obtain a target structure parameter.

Description

Technical Field

[0001] The present disclosure relates to the field of architectural design, and in particular to a method for determining structural parameters of a building structure and related equipment. Background Art

[0002] Traditional building structure design mainly relies on manual empirical formulas and simplified models. However, these methods are often inefficient when dealing with complex structural dynamic problems and do not take into account the impact of multi-factor interactions on vibration performance. For example, they ignore certain key dynamic loads, such as wind loads and earthquakes. Therefore, existing methods are difficult to accurately predict complex nonlinear dynamic loads and coupling effects, have high computational complexity, require a lot of computing resources and time, and rely on manual experience and experiments, resulting in low efficiency and accuracy in building structure design, affecting the practicality of engineering applications. Summary of the invention

[0003] The present disclosure proposes a method for determining structural parameters of a building structure and related equipment, so as to solve the technical problems of low efficiency and low accuracy in building structure design to a certain extent.

[0004] In a first aspect, the present disclosure provides a method for determining structural parameters of a building structure, comprising:

[0005] Acquiring first structural parameters and preset environmental parameters about the building structure;

[0006] Inputting the first structural parameter and the preset environmental parameter into a trained response prediction model to obtain a corresponding first response; wherein the response prediction model is trained based on static analysis results and dynamic analysis results of a building structure sample, the static analysis results include a static response of the building structure sample under a static load, and the dynamic analysis results include a vibration response of the building structure sample under a dynamic load;

[0007] The first structural parameters are updated based on the first response and a preset optimization target to obtain target structural parameters.

[0008] In some embodiments, the method further comprises:

[0009] Constructing a geometric model corresponding to the building structure sample based on the sample structure parameters of the building structure sample;

[0010] Performing analysis based on the geometric model, material properties corresponding to the building structure sample, environmental parameters and preset boundary conditions to obtain static analysis results and dynamic analysis results of the building structure sample;

[0011] The initial model is trained based on the static analysis result and the dynamic analysis result to obtain the trained response prediction model.

[0012] In some embodiments, the sample structure parameters include a geometric structure and a corresponding size, and equipment parameters, the geometric structure is used to indicate a shape structure constituting the building structure, the corresponding size is used to indicate a size of the shape structure, and the equipment parameters include equipment type, equipment size, and equipment location of equipment set in the building structure sample;

[0013] Then, a geometric model corresponding to the building structure sample is constructed based on the sample structure parameters of the building structure sample, including:

[0014] Constructing a three-dimensional model of the building structure sample based on the geometric structure and corresponding dimensions of the building structure sample;

[0015] The corresponding device model is set in the three-dimensional model based on the device parameters to obtain the geometric model.

[0016] In some embodiments, the analysis is performed according to the geometric model, the material properties corresponding to the building structure sample, the environmental parameters and the preset boundary conditions to obtain the static analysis results and the dynamic analysis results of the building structure sample, including:

[0017] Meshing the geometric model to obtain a finite element model;

[0018] Calculating the stress, strain and displacement of the preset key structural parts of the building structure sample of the finite element model under static load based on the material properties and the preset boundary conditions to obtain the static analysis results;

[0019] The vibration response of the preset key structural parts of the building structure sample under the dynamic load of the finite element model is calculated based on the material properties, the environmental parameters and the preset boundary conditions to obtain the dynamic analysis results.

[0020] In some embodiments, the initial model includes stacked multi-layer restricted Boltzmann machines;

[0021] Then, the initial model is trained based on the static analysis result and the dynamic analysis result to obtain the trained response prediction model, including:

[0022] Using the sample structure parameters of the building structure sample as input layer data, using the static analysis results and the dynamic analysis results as output layer data, training each layer of restricted Boltzmann machine layer by layer based on contrast divergence from the bottom layer to obtain the hierarchical training weights of each layer of restricted Boltzmann machine; wherein the hierarchical training weights of each layer of restricted Boltzmann machine are used as the initialization weights of the next layer of restricted Boltzmann machine;

[0023] The hierarchical training weight progress is adjusted based on the back propagation algorithm, the target weight of the restricted Boltzmann machine of each layer is determined, and the trained response prediction model is obtained.

[0024] In some embodiments, adjusting the hierarchical training weight progress based on a back propagation algorithm to determine the target weight of each layer of the restricted Boltzmann machine includes:

[0025] w′ is the target weight, w is the layer training weight, η is the learning rate, and loss(y, t) is the loss function between the model prediction value y and the output layer data t.

[0026] In some embodiments, updating the first structural parameter based on the first response and a preset optimization target to obtain a target structural parameter includes:

[0027] In response to the first response not satisfying the preset optimization target, updating the first structural parameter based on the first response to obtain a second structural parameter;

[0028] Inputting the second structural parameter into the response prediction model to obtain a corresponding second response;

[0029] Determining whether the second response meets the preset optimization goal;

[0030] In response to the second response satisfying the preset optimization target, determining the second structural parameter as the target structural parameter;

[0031] In response to the second response not satisfying the preset optimization target, the second structural parameter is updated based on the second response.

[0032] In a second aspect of the present disclosure, a device for determining structural parameters of a building structure is provided, comprising:

[0033] An acquisition module, used for acquiring first structural parameters and preset environmental parameters about the building structure;

[0034] A prediction module, used for inputting the first structural parameter and the preset environmental parameter into a trained response prediction model to obtain a corresponding first response; wherein the response prediction model is trained based on static analysis results and dynamic analysis results of a building structure sample, the static analysis results include a static response of the building structure sample under a static load, and the dynamic analysis results include a vibration response of the building structure sample under a dynamic load;

[0035] A parameter determination module is used to update the first structural parameters based on the first response and a preset optimization target to obtain target structural parameters.

[0036] In some embodiments, the apparatus further comprises:

[0037] A model module, used for constructing a geometric model corresponding to the building structure sample based on the sample structure parameters of the building structure sample;

[0038] An analysis module, used to perform analysis according to the geometric model, material properties corresponding to the building structure sample, environmental parameters and preset boundary conditions, to obtain static analysis results and dynamic analysis results of the building structure sample;

[0039] The training module is used to train the initial model based on the static analysis results and the dynamic analysis results to obtain the trained response prediction model.

[0040] In some embodiments, the sample structure parameters include a geometric structure and a corresponding size, and equipment parameters, the geometric structure is used to indicate a shape structure constituting the building structure, the corresponding size is used to indicate a size of the shape structure, and the equipment parameters include equipment type, equipment size, and equipment location of equipment set in the building structure sample;

[0041] Then, a geometric model corresponding to the building structure sample is constructed based on the sample structure parameters of the building structure sample, including:

[0042] Constructing a three-dimensional model of the building structure sample based on the geometric structure and corresponding dimensions of the building structure sample;

[0043] The corresponding device model is set in the three-dimensional model based on the device parameters to obtain the geometric model.

[0044] In some embodiments, the analysis is performed according to the geometric model, the material properties corresponding to the building structure sample, the environmental parameters and the preset boundary conditions to obtain the static analysis results and the dynamic analysis results of the building structure sample, including:

[0045] Meshing the geometric model to obtain a finite element model;

[0046] Calculating the stress, strain and displacement of the preset key structural parts of the building structure sample of the finite element model under static load based on the material properties and the preset boundary conditions to obtain the static analysis results;

[0047] The vibration response of the preset key structural parts of the building structure sample under the dynamic load of the finite element model is calculated based on the material properties, the environmental parameters and the preset boundary conditions to obtain the dynamic analysis results.

[0048] In some embodiments, the initial model includes stacked multi-layer restricted Boltzmann machines;

[0049] Then, the initial model is trained based on the static analysis result and the dynamic analysis result to obtain the trained response prediction model, including:

[0050] Using the sample structure parameters of the building structure sample as input layer data, using the static analysis results and the dynamic analysis results as output layer data, training each layer of restricted Boltzmann machine layer by layer based on contrast divergence from the bottom layer to obtain the hierarchical training weights of each layer of restricted Boltzmann machine; wherein the hierarchical training weights of each layer of restricted Boltzmann machine are used as the initialization weights of the next layer of restricted Boltzmann machine;

[0051] The hierarchical training weight progress is adjusted based on the back propagation algorithm, the target weight of the restricted Boltzmann machine of each layer is determined, and the trained response prediction model is obtained.

[0052] In some embodiments, adjusting the hierarchical training weight progress based on a back propagation algorithm to determine the target weight of each layer of the restricted Boltzmann machine includes:

[0053] w′ is the target weight, w is the layer training weight, η is the learning rate, and loss(y, t) is the loss function between the model prediction value y and the output layer data t.

[0054] In some embodiments, updating the first structural parameter based on the first response and a preset optimization target to obtain a target structural parameter includes:

[0055] In response to the first response not satisfying the preset optimization target, updating the first structural parameter based on the first response to obtain a second structural parameter;

[0056] Inputting the second structural parameter into the response prediction model to obtain a corresponding second response;

[0057] Determining whether the second response meets the preset optimization goal;

[0058] In response to the second response satisfying the preset optimization target, determining the second structural parameter as the target structural parameter;

[0059] In response to the second response not satisfying the preset optimization target, the second structural parameter is updated based on the second response.

[0060] In a third aspect of the present disclosure, an electronic device is provided, comprising one or more processors, a memory; and one or more programs, wherein the one or more programs are stored in the memory and executed by the one or more processors, and the programs include instructions for executing the method described in the first aspect.

[0061] According to a fourth aspect of the present disclosure, a non-volatile computer-readable storage medium containing a computer program is provided. When the computer program is executed by one or more processors, the processors execute the method described in the first aspect.

[0062] In a fifth aspect of the present disclosure, a computer program product is provided, comprising computer program instructions, which, when executed on a computer, enable the computer to execute the method described in the first aspect.

[0063] From the above description, it can be seen that the present invention provides a method for determining structural parameters of a building structure and related equipment, which reduces the complexity of building structure design, improves the efficiency and accuracy of building structure design, and saves time and cost of experiments and manual analysis by performing predictions through a trained neural network model and adjusting design parameters according to the prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the technical solutions in the present disclosure or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0065] Figure 1 A schematic diagram of a framework for determining structural parameters of a building structure according to an embodiment of the present disclosure.

[0066] Figure 2 The figure is a schematic diagram of the hardware structure of an exemplary electronic device according to an embodiment of the present disclosure.

[0067] Figure 3 It is a flowchart of a method for determining structural parameters of a building structure according to an embodiment of the present disclosure.

[0068] Figure 4It is a schematic diagram of a device for determining structural parameters of a building structure according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0069] In order to make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the present disclosure is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0070] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present disclosure should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the embodiments of the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0071] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0072] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0073] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet the relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0074] Figure 1 A schematic diagram of a structural parameter determination framework of a building structure according to an embodiment of the present disclosure is shown. Figure 1The structural parameter determination architecture 100 of the building structure may include a server 110, a terminal 120, and a network 130 providing a communication link. The server 110 and the terminal 120 may be connected via a wired or wireless network 130. The server 110 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, security services, and CDN.

[0075] The terminal 120 may be implemented in hardware or software. For example, when the terminal 120 is implemented in hardware, it may be various electronic devices having a display screen and supporting page display, including but not limited to smart phones, tablet computers, e-book readers, laptop portable computers, and desktop computers, etc. When the terminal 120 device is implemented in software, it may be installed in the electronic devices listed above; it may be implemented as multiple software or software modules (such as software or software modules used to provide distributed services), or it may be implemented as a single software or software module, which is not specifically limited here.

[0076] It should be noted that the method for determining the structural parameters of the building structure provided in the embodiment of the present application can be executed by the terminal 120 or by the server 110. It should be understood that Figure 1 The number of terminals, networks and servers in the embodiment is only for illustration and is not intended to limit the number of terminals, networks and servers. Any number of terminals, networks and servers may be provided as required.

[0077] Figure 2 FIG. 2 shows a schematic diagram of the hardware structure of an exemplary electronic device 200 provided in an embodiment of the present disclosure. Figure 2 As shown, the electronic device 200 may include: a processor 202, a memory 204, a network module 206, a peripheral interface 208 and a bus 210. The processor 202, the memory 204, the network module 206 and the peripheral interface 208 are connected to each other in communication within the electronic device 200 through the bus 210.

[0078] Processor 202 may be a central processing unit (CPU), a structural parameter determiner of a building structure, a neural network processor (NPU), a microcontroller (MCU), a programmable logic device, a digital signal processor (DSP), an application specific integrated circuit (ASIC), or one or more integrated circuits. Processor 202 may be used to perform functions related to the technology described in the present disclosure. In some embodiments, processor 202 may also include multiple processors integrated into a single logical component. For example, Figure 2 As shown, processor 202 may include a plurality of processors 202a, 202b, and 202c.

[0079] The memory 204 may be configured to store data (eg, instructions, computer code, etc.). Figure 2 As shown, the data stored in the memory 204 may include program instructions (for example, program instructions for implementing the method for determining the structural parameters of the building structure of the embodiment of the present disclosure) and data to be processed (for example, the memory may store configuration files of other modules, etc.). The processor 202 may also access the program instructions and data stored in the memory 204, and execute the program instructions to operate on the data to be processed. The memory 204 may include a volatile storage device or a non-volatile storage device. In some embodiments, the memory 204 may include a random access memory (RAM), a read-only memory (ROM), an optical disk, a magnetic disk, a hard disk, a solid-state drive (SSD), a flash memory, a memory stick, etc.

[0080] The network module 206 can be configured to provide communication with other external devices to the electronic device 200 via a network. The network can be any wired or wireless network capable of transmitting and receiving data. For example, the network can be a wired network, a local wireless network (e.g., Bluetooth, WiFi, near field communication (NFC) etc.), a cellular network, the Internet or a combination thereof. It is understood that the type of network is not limited to the above specific examples. In some embodiments, the network module 306 can include any number of network interface controllers (NICs), radio frequency modules, transceivers, modems, routers, gateways, adapters, cellular network chips, etc., in any combination.

[0081] The peripheral interface 208 can be configured to connect the electronic device 200 to one or more peripheral devices to achieve information input and output. For example, the peripheral devices can include input devices such as a keyboard, a mouse, a touch pad, a touch screen, a microphone, and various sensors, and output devices such as a display, a speaker, a vibrator, and an indicator light.

[0082] The bus 210 can be configured to transmit information between various components of the electronic device 200 (e.g., the processor 202, the memory 204, the network module 206, and the peripheral interface 208), such as an internal bus (e.g., a processor-memory bus), an external bus (USB port, PCI-E bus), etc.

[0083] It should be noted that, although the architecture of the electronic device 200 only shows the processor 202, the memory 204, the network module 206, the peripheral interface 208 and the bus 210, in the specific implementation process, the architecture of the electronic device 200 may also include other components necessary for normal execution. In addition, it can be understood by those skilled in the art that the architecture of the electronic device 200 may also only include the components necessary for implementing the embodiments of the present disclosure, and does not necessarily include all the components shown in the figure.

[0084] Traditional structural vibration analysis of building structures mainly relies on artificial empirical formulas and simplified models. However, these methods are often inefficient when dealing with complex structural dynamic problems, and do not take into account the impact of multi-factor interactions on vibration performance, especially in the dynamic performance analysis of large structures. This problem is more prominent. The existing vibration calculation and parameter design of building structures often ignore certain key dynamic loads, such as wind loads and earthquakes. In addition, it is difficult for existing methods to accurately predict complex nonlinear dynamic loads and coupling effects. In addition, traditional analysis also fails to effectively consider the impact of material properties and boundary conditions on vibration performance. Therefore, the existing building structure design calculation is highly complex and requires a lot of computing resources and time. Moreover, it is impossible to accurately simulate the nonlinear dynamic response of the actual structure, and often ignores the influence of factors such as site conditions and working conditions, such as earthquakes, temperature, humidity changes, etc., resulting in inaccurate design results and affecting the practicality of engineering applications. It also relies on manual experience and experiments, resulting in low efficiency of building design. Therefore, how to reduce the complexity of building structure design and improve the efficiency and accuracy of building structure design has become a technical problem that needs to be solved urgently.

[0085] In view of this, the embodiments of the present disclosure provide a method for determining structural parameters of a building structure and related equipment. By predicting through a trained neural network model and adjusting the design parameters according to the prediction results, the complexity of the building structure design is reduced, the efficiency and accuracy of the building structure design are improved, and the time and cost of experiments and manual analysis are saved.

[0086] See also Figure 3 , Figure 3 A schematic flow chart of a method for determining structural parameters of a building structure according to an embodiment of the present disclosure is shown. The method for determining structural parameters of a building structure according to an embodiment of the present disclosure can be deployed on a server or client. Figure 3In the method 300 for determining structural parameters of a building structure, the method 300 may further include the following steps.

[0087] In step S310, first structural parameters and preset environmental parameters about the building structure are obtained.

[0088] Among them, the structural parameters of the building structure can include the description parameters of the building shape, spatial layout and its components. The structural parameters are not only the embodiment of the appearance of the building structure, but also contain all the geometric information required for structural mechanics analysis, so that the performance indicators such as force, deformation and stability of the structure under various loads can be calculated. The structural parameters of the building structure can include: Nodes, which are used to represent the locations where structural elements intersect or connect in space. Elements, which are used to connect line segments or patches between nodes to simulate beams, columns, plates and other components in actual structures. For example, rods, beams, shells, solids, etc., each type corresponds to different physical and mathematical models. Boundary conditions, which are used to describe the connection between the structure and the outside world, such as fixed ends, hinged ends, free ends, etc., as well as external constraints such as supports and supports. Material properties, which can refer to the material properties assigned to each element in the model, such as elastic modulus, Poisson's ratio, density, etc., which determine the behavior of the material when subjected to force. The structural parameters of the building structure may also include equipment parameters, which may include parameters such as equipment type, equipment size, equipment location, etc. of the equipment arranged in the building structure.

[0089] It should be understood that the first structural parameter may refer to the structural parameter that needs to be optimized currently, which may be an initial structural parameter or an intermediate construction parameter in the optimization iteration process. When the first structural parameter is an intermediate structural parameter in the iteration process, it may represent the current structural parameter in the current round of iteration.

[0090] Environmental parameters may refer to parameters that affect the performance, safety and durability of building structures, and may include at least one of the following: climate parameters (e.g., temperature, humidity, wind speed and wind speed, precipitation, etc.), geological and topographic parameters (e.g., soil type and properties, terrain slope, earthquake intensity), environmental load parameters (e.g., snow load, ice load, flood risk, etc.), and man-made environmental parameters (e.g., pollution and corrosion parameters, traffic vibration parameters, etc.). The preset environmental parameters may be obtained based on actual environmental detection of the building structure.

[0091] In step S320, the first structural parameter and the preset environmental parameter are input into a trained response prediction model to obtain a corresponding first response; wherein the response prediction model is trained based on static analysis results and dynamic analysis results of the building structure sample, the static analysis result includes the static response of the building structure sample under static load, and the dynamic analysis result includes the vibration response of the building structure sample under dynamic load.

[0092] Among them, the trained response prediction model is used to predict the response of the building structure with the first structural parameters in the environment of the preset environmental parameters, for example, it can include static response prediction and dynamic response prediction. The static analysis results can represent the static response of the building structure under the action of constant or slowly changing external loads. For the pumped storage power plant building, this includes the deadweight (the weight of the building's own materials), the weight of fixed equipment, wind pressure (considering the lateral force on the structure under the maximum wind speed), etc. The static analysis results can determine the structural stress distribution, strain (degree of deformation) and displacement (the movement of each point of the structure relative to the initial position) under these long-term forces to verify whether the strength and stability of the structure meet the design specification requirements. For example, considering the roof design of the pumped storage power plant building, static analysis can be used to evaluate whether the beams and columns of the roof structure will have excessive stress concentration under the action of deadweight and expected maximum snow load (if applicable), resulting in material yield or fracture, and whether the maximum deflection of the roof (the degree of downward bending) is within an acceptable range to ensure that it will not affect normal use or aesthetics.

[0093] The results of dynamic analysis can represent the dynamic response of the building structure under the external loads that change rapidly over time. External loads that change rapidly over time may cause structural vibrations. For pumped storage power plant buildings, dynamic loads can include mechanical vibrations (such as vibrations generated by the operation of generator sets), hydraulic pulsation pressure (water pressure fluctuations caused by rapid changes in water flow during pumped storage), and earthquakes. Dynamic analysis structures can represent the vibration characteristics of building structures, such as natural frequency, mode shape (the vibration form of the structure at a specific frequency), and response under specific excitation, including displacement, velocity, acceleration, etc., to ensure that the structure remains safe and stable under dynamic loads. For example, in a pumped storage power plant building, the pump-turbine will generate significant mechanical vibrations during the conversion process. The dynamic analysis results will simulate the impact of these vibrations on the foundation and connection structure of the power plant, determine whether there is a risk of resonance, and whether the damping effect (energy dissipation capacity) of the structure is sufficient to avoid excessive vibration. In addition, considering that earthquakes are sudden dynamic loads, dynamic analysis can also evaluate the seismic performance of the power plant under specific earthquake intensities to ensure that the structure can withstand the expected maximum earthquake force without collapse or serious damage.

[0094] In some embodiments, method 300 further includes:

[0095] Constructing a geometric model corresponding to the building structure sample based on the sample structure parameters of the building structure sample;

[0096] Performing analysis based on the geometric model, material properties corresponding to the building structure sample, environmental parameters and preset boundary conditions to obtain static analysis results and dynamic analysis results of the building structure sample;

[0097] The initial model is trained based on the static analysis result and the dynamic analysis result to obtain the trained response prediction model.

[0098] Among them, using static analysis results and dynamic analysis results to train the response prediction model can enable the response prediction model to explore the nonlinear intrinsic relationship between the structural parameters, environmental parameters and corresponding responses of the building structure, so as to accurately predict the corresponding static response and dynamic response based on the given structural parameters and environmental parameters.

[0099] In some embodiments, the sample structure parameters include a geometric structure and a corresponding size, and equipment parameters, the geometric structure is used to indicate a shape structure constituting the building structure, the corresponding size is used to indicate a size of the shape structure, and the equipment parameters include equipment type, equipment size, and equipment location of equipment set in the building structure sample;

[0100] Then, a geometric model corresponding to the building structure sample is constructed based on the sample structure parameters of the building structure sample, including:

[0101] Constructing a three-dimensional model of the building structure sample based on the geometric structure and corresponding dimensions of the building structure sample;

[0102] The corresponding device model is set in the three-dimensional model based on the device parameters to obtain the geometric model.

[0103] Specifically, taking the pumped storage power plant as an example, the finite element model of the pumped storage power plant can be established using structural analysis software, including the main structure of the plant and related environmental parameters. First, determine the actual structure of the pumped storage power plant to be simulated, including all parts that need to be considered, such as the main body of the plant, machinery and equipment, and infrastructure. In the structural analysis software (such as ANSYS or Abaqus), import the existing three-dimensional model, or create a three-dimensional geometric model according to the actual size and structure. Perform plane or volume meshing on the three-dimensional geometric model to obtain a finite element model. The unit type and size can be reasonably selected as needed to ensure calculation accuracy and efficiency. Then, the corresponding material properties, such as density, elastic modulus, Poisson's ratio, etc., can be assigned to the geometric model. And the boundary conditions of the model, such as fixed constraints, sliding constraints, etc., can be set according to actual conditions. At the same time, simulate the actual loads that may be borne, including deadweight, hydraulic pulsation pressure, mechanical vibration, earthquake, etc. Set relevant environmental parameters according to actual working conditions, topography, climate and other factors. Finally, check whether the model settings are correct, and the correctness of the model can be confirmed by comparing the calculation results of the preset problems. In the geometric model of the pumped storage power plant building, nodes can be located at key intersections such as the top of the column, the connection point of the beam, and the corner of the wall. For example, the point where the bottom of each column of the plant is connected to the foundation is defined as a node, and the point where the top of the column connects to the beam is also a node. Beam elements can be used to simulate the beams and longitudinal beams of the plant, bear the loads of the roof and floor, and transmit dynamic loads such as mechanical vibration. Column elements can represent supporting structures, bearing vertical loads and horizontal forces, such as deadweight, equipment weight, wind loads, and seismic forces. Plate and shell elements can simulate roof and floor structures, with special attention to their ability to withstand uniformly distributed loads (such as snow loads) and dynamic water pressure (such as when the reservoir is overflowing). Wall elements can simulate the maintenance structure around the plant, considering its effect on the overall stiffness and stability, as well as functions such as sound insulation and waterproofing.

[0104] In some embodiments, the analysis is performed according to the geometric model, the material properties corresponding to the building structure sample, the environmental parameters and the preset boundary conditions to obtain the static analysis results and the dynamic analysis results of the building structure sample, including:

[0105] Meshing the geometric model to obtain a finite element model;

[0106] Calculating the stress, strain and displacement of the finite element model under static load based on the material properties and the preset boundary conditions to obtain the static analysis result;

[0107] The vibration response of the finite element model under dynamic load is calculated based on the material properties, the environmental parameters and the preset boundary conditions to obtain the dynamic analysis result.

[0108] Among them, after setting the boundary conditions and material properties, the finite element method can be used to perform static and dynamic analysis on the pumped storage power station plant model. Static analysis mainly solves the model's stress, strain, displacement and other parameters under static loads (such as deadweight, wind pressure, etc.). Dynamic analysis examines the response of the plant under dynamic loads (such as hydraulic pulsation pressure, mechanical vibration, earthquake, etc.), including vibration mode, frequency response and other parameters.

[0109] Specifically, the preset boundary conditions may include fixed supports (completely constraining the movement and rotation of nodes), sliding supports (allowing the structure to move freely in one or several directions), etc., which are usually found in pumped storage power station buildings. For example, the nodes of the foundation part are usually set as fixed supports to simulate their firm connection with the ground; while some specially designed connection nodes may allow a certain displacement to adapt to foundation settlement or thermal expansion and contraction. Material properties can be assigned corresponding elastic modulus, Poisson's ratio, density, etc. to each element based on the materials to be used, such as concrete, steel, etc., to reflect the response characteristics of the material under stress. Based on the above data, the performance of the building structure under different working conditions (such as normal operation, full-load pumped storage, earthquakes and other extreme events) can be simulated, including stress distribution, deformation, dynamic response, etc., to ensure that the design meets the requirements of safety and durability. For example, through dynamic analysis, the seismic performance of the plant under earthquake can be evaluated to ensure that the structure is not damaged under the design earthquake intensity, and to ensure the safety of personnel and the normal operation of equipment. At the same time, static analysis helps to verify the strength and stability of the structure under long-term static loads to ensure the reliability of the structure during long-term service.

[0110] In some embodiments, the initial model includes stacked multi-layer restricted Boltzmann machines;

[0111] Then, the initial model is trained based on the static analysis result and the dynamic analysis result to obtain the trained response prediction model, including:

[0112] Using the sample structure parameters of the building structure sample as input layer data, using the static analysis results and the dynamic analysis results as output layer data, training each layer of restricted Boltzmann machine layer by layer based on contrast divergence from the bottom layer to obtain the hierarchical training weights of each layer of restricted Boltzmann machine; wherein the hierarchical training weights of each layer of restricted Boltzmann machine are used as the initialization weights of the next layer of restricted Boltzmann machine;

[0113] The hierarchical training weight progress is adjusted based on the back propagation algorithm, the target weight of the restricted Boltzmann machine of each layer is determined, and the trained response prediction model is obtained.

[0114] Among them, the structural dimensions and material properties can be used as input parameters of the response prediction model. For example, the structural dimensions determine the geometric characteristics of the model, and the material properties (such as density, elastic modulus, Poisson's ratio, etc.) affect the mechanical behavior of the model. The output parameters of the response prediction model can be static response and / or dynamic response, and can also include performance indicators. The static analysis results and dynamic analysis results obtained based on the geometric model analysis can be used as sample data to create and train a neural network model for learning and simulating the complex nonlinear relationship between input parameters and output responses. Specifically, a deep belief network (DBN) can be used, which can include a stacked combination of multiple layers of restricted Boltzmann machines (RBMs), and the hidden layer of each layer of RBM is used as the input of the next layer, stacked layer by layer, and the layers are connected into a tree structure. As the basic building block of DBN, RBM is a two-layer neural network, including a visible layer (usually corresponding to the input data) and a hidden layer. The neurons between layers are fully connected, but there is no connection within the same layer. By learning weights through algorithms such as Contrastive Divergence, it is possible to capture high-order statistical features of the data. DBN forms a deep structure by stacking multiple RBMs, and the hidden layer output of each layer serves as the input of the next layer. The top-level RBM can be an ordinary Boltzmann machine because it has no further hidden layers. In DBN, it is generally assumed that each layer is binary (0 or 1) data; let the state vector of the lth layer be h (l) , where L = 0 represents the input layer, L represents the last hidden layer, and when L = 2, it is the simplest single hidden layer RBM model, i.e. h (l) =h,h (0) =v,h (2) =h′,h (l) The probability distribution of can be defined by the state of the next layer, for l<L:

[0115] For the top layer L, its probability is determined by itself:

[0116] Among them, σ represents the activation function; in the undirected RBM, the connection between any two layers is symmetric, that is, In practical applications, the top layer L of DBN is taken as a feature; the data of the input layer is taken as a feature for supervised learning; the input data, the physical parameters of the plant structure, the earthquake intensity, and the mechanical properties of the building materials are input into the input layer of DBN; and the vibration response results of the plant structure, such as the vibration amplitude and frequency, are taken as labels for supervised learning; for DBN, the data of the input layer is regarded as a feature for training and adjusting the weights and biases of the network; and the data of the output layer, that is, the label, is the target value that the network wants to predict.

[0117] The deep belief network DBN can be trained by pre-training the layers first. This is an unsupervised training process. On each layer, a restricted Boltzmann machine RBM is trained first. This process can be achieved by a greedy hierarchical pre-training algorithm, using the hidden state of the previous layer as the input of the next layer. The training process of RBM can be carried out by contrastive divergence (, and the contrastive divergence update formula is as follows:

[0118] The change in weight Δw ij =ε( <v i h j > data - <v i h j > model ), where <v i h j > data is the statistical expectation from the data distribution, <v i h j > model is the statistical expectation from the model distribution, and ε is the learning rate.

[0119] In some embodiments, adjusting the hierarchical training weight progress based on a back propagation algorithm to determine the target weight of each layer of the restricted Boltzmann machine includes:

[0120] w′ is the target weight, w is the layer training weight, η is the learning rate, and loss(y, t) is the loss function between the model prediction value y and the output layer data t.

[0121] After pre-training each layer of RBM, the entire DBN is fine-tuned as a deep network in a supervised manner. The back-propagation algorithm can be used to adjust the weights of all layers to optimize the final classification or regression task. In this process, DBN is regarded as a multi-layer neural network and trained through the back-propagation algorithm. The weights learned in the pre-training process are adjusted based on quadratic optimization to enable the network to model the prediction target more accurately. The gradient adjustment formula is as follows:

[0122] Wherein, w′ is the target weight, w is the layer training weight, η is the learning rate, and loss(y, t) is the loss function between the model prediction value y and the output layer data t. For example, loss(y, t) can be a cross entropy loss function.

[0123] According to the response prediction model of the embodiment of the present disclosure, a DBN model is adopted, which has multiple hidden layers, and can better capture and model the high-dimensional complexity of the data, so that the complex nonlinear relationship between the input parameters and the output response can be learned and modeled more effectively. In DBN, pre-training is performed in an unsupervised manner, and useful features can be automatically learned and extracted from the training data, which is extremely beneficial for complex or incomprehensible data (such as plant structure vibration data). Through unsupervised pre-training, DBN can obtain a relatively optimized initial weight, which helps to avoid the gradient vanishing problem that usually exists in deep neural networks, and also helps the model converge faster. After pre-training, DBN also performs a back-propagation fine-tuning process so that the model can better adapt to the training data through supervised learning while maintaining the learned characteristics. In this way, when processing finite element analysis data, DBN can not only simulate known input and output relationships, but also learn and simulate unknown or hidden relationships. DBN is modular and can adapt to problems of different complexities by adding more hidden layers or changing the number of nodes in each layer. This flexibility and scalability make DBN very suitable for handling complex plant structure vibration simulation problems.

[0124] In step S330, the first structural parameters are updated based on the first response and a preset optimization target to obtain target structural parameters.

[0125] The preset optimization target may refer to a preset condition that the building structure needs to meet, such as maximizing or minimizing a specific parameter (e.g., minimizing the vibration of the structure), or satisfying certain specific constraints (e.g., maintaining the stability of the structure or satisfying specific design specifications). When the first response meets the preset optimization target, the first structural parameter may be directly determined as the target structural parameter. When the first response does not meet the preset optimization target, the first structural parameter is adjusted and iterated until the response corresponding to the structural parameter meets the preset condition, and the corresponding structural parameter is the target structural parameter that meets the preset condition.

[0126] In some embodiments, updating the first structural parameter based on the first response to obtain a target structural parameter includes:

[0127] In response to the first response not satisfying the preset optimization target, updating the first structural parameter based on the first response to obtain a second structural parameter;

[0128] Inputting the second structural parameter into the response prediction model to obtain a corresponding second response;

[0129] Determining whether the second response meets the preset optimization goal;

[0130] In response to the second response satisfying the preset optimization target, determining the second structural parameter as the target structural parameter;

[0131] In response to the second response not satisfying the preset optimization target, the second structural parameter is updated based on the second response.

[0132] Among them, the second structural parameter may refer to the structural parameter after the current structural parameter is updated in the current round of iteration. When the response corresponding to the updated structural parameter meets the preset requirements, the updated structural parameter is the target structural parameter. When the response corresponding to the updated structural parameter still does not meet the preset requirements, continue to update the updated structural parameter until the updated structural parameter can meet the preset optimization target. For example, a preset optimization target can be determined, and the structural parameters in the initial design (such as the size of the plant, the layout of the supporting structure, the materials used, etc.) are used as inputs of the response prediction model. Using the trained response prediction model, the corresponding structural vibration response is predicted according to the given input parameters. According to the prediction results, the input building structure design parameters are adjusted to achieve the optimization target. Generally, an optimization algorithm (such as gradient descent method, genetic algorithm, etc.) can be used for system optimization. The above process is repeatedly performed until the preset optimization target is reached, or the optimization result has met the preset conditions.

[0133] It can be seen that according to the method of the embodiment of the present disclosure, a finite element model is constructed by utilizing structural analysis, and the finite element method is used to perform static and dynamic analysis on the model after setting boundary conditions and material properties. Furthermore, appropriate input and output parameters are selected, and a neural network model is used for training and simulation to achieve structural optimization in early design, effectively improving the accuracy and efficiency of optimization. This method achieves modeling of more complex nonlinear relationships by introducing a deep confidence network model, and optimizes the network model through a supervised learning method, further improving the accuracy of vibration simulation. At the same time, the present invention fully considers the influence of various working conditions, making the prediction results more accurate, greatly saving the time and cost of experiments and manual analysis, and improving the accuracy of the design.

[0134] It should be noted that the method of the embodiment of the present disclosure can be performed by a single device, such as a computer or a server. The method of the present embodiment can also be applied in a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present disclosure, and the multiple devices will interact with each other to complete the described method.

[0135] It should be noted that the above describes some embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0136] Based on the same technical concept, corresponding to any of the above-mentioned embodiments and methods, the present disclosure also provides a device for determining structural parameters of a building structure, see Figure 4 , the structural parameter determination device of the building structure comprises:

[0137] An acquisition module, used for acquiring first structural parameters and preset environmental parameters about the building structure;

[0138] A prediction module, used for inputting the first structural parameter and the preset environmental parameter into a trained response prediction model to obtain a corresponding first response; wherein the response prediction model is trained based on static analysis results and dynamic analysis results of a building structure sample, the static analysis results include a static response of the building structure sample under a static load, and the dynamic analysis results include a vibration response of the building structure sample under a dynamic load;

[0139] A parameter determination module is used to update the first structural parameters based on the first response and a preset optimization target to obtain target structural parameters.

[0140] In some embodiments, the apparatus further comprises:

[0141] A model module, used for constructing a geometric model corresponding to the building structure sample based on the sample structure parameters of the building structure sample;

[0142] An analysis module, used to perform analysis according to the geometric model, material properties corresponding to the building structure sample, environmental parameters and preset boundary conditions, to obtain static analysis results and dynamic analysis results of the building structure sample;

[0143] The training module is used to train the initial model based on the static analysis results and the dynamic analysis results to obtain the trained response prediction model.

[0144] In some embodiments, the sample structure parameters include a geometric structure and a corresponding size, and equipment parameters, the geometric structure is used to indicate a shape structure constituting the building structure, the corresponding size is used to indicate a size of the shape structure, and the equipment parameters include equipment type, equipment size, and equipment location of equipment set in the building structure sample;

[0145] Then, a geometric model corresponding to the building structure sample is constructed based on the sample structure parameters of the building structure sample, including:

[0146] Constructing a three-dimensional model of the building structure sample based on the geometric structure and corresponding dimensions of the building structure sample;

[0147] The corresponding device model is set in the three-dimensional model based on the device parameters to obtain the geometric model.

[0148] In some embodiments, the analysis is performed according to the geometric model, the material properties corresponding to the building structure sample, the environmental parameters and the preset boundary conditions to obtain the static analysis results and the dynamic analysis results of the building structure sample, including:

[0149] Meshing the geometric model to obtain a finite element model;

[0150] Calculating the stress, strain and displacement of the preset key structural parts of the building structure sample of the finite element model under static load based on the material properties and the preset boundary conditions to obtain the static analysis results;

[0151] The vibration response of the preset key structural parts of the building structure sample under the dynamic load of the finite element model is calculated based on the material properties, the environmental parameters and the preset boundary conditions to obtain the dynamic analysis results.

[0152] In some embodiments, the initial model includes stacked multi-layer restricted Boltzmann machines;

[0153] Then, the initial model is trained based on the static analysis result and the dynamic analysis result to obtain the trained response prediction model, including:

[0154] Using the sample structure parameters of the building structure sample as input layer data, using the static analysis results and the dynamic analysis results as output layer data, training each layer of restricted Boltzmann machine layer by layer based on contrast divergence from the bottom layer to obtain the hierarchical training weights of each layer of restricted Boltzmann machine; wherein the hierarchical training weights of each layer of restricted Boltzmann machine are used as the initialization weights of the next layer of restricted Boltzmann machine;

[0155] The hierarchical training weight progress is adjusted based on the back propagation algorithm, the target weight of the restricted Boltzmann machine of each layer is determined, and the trained response prediction model is obtained.

[0156] In some embodiments, adjusting the hierarchical training weight progress based on a back propagation algorithm to determine the target weight of each layer of the restricted Boltzmann machine includes:

[0157] w′ is the target weight, w is the layer training weight, η is the learning rate, and loss(y, t) is the loss function between the model prediction value y and the output layer data t.

[0158] In some embodiments, updating the first structural parameter based on the first response and a preset optimization target to obtain a target structural parameter includes:

[0159] In response to the first response not satisfying the preset optimization target, updating the first structural parameter based on the first response to obtain a second structural parameter;

[0160] Inputting the second structural parameter into the response prediction model to obtain a corresponding second response;

[0161] Determining whether the second response meets the preset optimization goal;

[0162] In response to the second response satisfying the preset optimization target, determining the second structural parameter as the target structural parameter;

[0163] In response to the second response not satisfying the preset optimization target, the second structural parameter is updated based on the second response.

[0164] For the convenience of description, the above device is described by dividing it into various modules according to its functions. Of course, when implementing the present disclosure, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0165] The device of the above embodiment is used to implement the structural parameter determination method of the corresponding building structure in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0166] Based on the same technical concept, corresponding to any of the above-mentioned embodiment methods, the present disclosure also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the method for determining the structural parameters of the building structure as described in any of the above embodiments.

[0167] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0168] The computer instructions stored in the storage medium of the above embodiments are used to enable the computer to execute the method for determining the structural parameters of the building structure as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0169] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present disclosure (including the claims) is limited to these examples. Based on the concept of the present disclosure, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present disclosure as described above, which are not provided in detail for the sake of simplicity.

[0170] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present disclosure difficult to understand, the known power / ground connections to the integrated circuit (IC) chips and other components may or may not be shown in the provided figures. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present disclosure difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform on which the embodiments of the present disclosure will be implemented (that is, these details should be fully within the scope of understanding of those skilled in the art). Where specific details (e.g., circuits) are set forth to describe exemplary embodiments of the present disclosure, it is apparent to those skilled in the art that the embodiments of the present disclosure can be implemented without these specific details or with changes in these specific details. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0171] Although the present disclosure has been described in conjunction with specific embodiments of the present disclosure, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0172] The embodiments of the present disclosure are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A method for determining structural parameters of a building structure, characterized in that: include: Acquiring first structural parameters and preset environmental parameters about the building structure; Inputting the first structural parameter and the preset environmental parameter into a trained response prediction model to obtain a corresponding first response; wherein the response prediction model is trained based on static analysis results and dynamic analysis results of a building structure sample, the static analysis results include a static response of the building structure sample under a static load, and the dynamic analysis results include a vibration response of the building structure sample under a dynamic load; The first structural parameters are updated based on the first response and a preset optimization target to obtain target structural parameters.

2. The method according to claim 1, characterized in that Also includes: Constructing a geometric model corresponding to the building structure sample based on the sample structure parameters of the building structure sample; Performing analysis based on the geometric model, material properties corresponding to the building structure sample, environmental parameters and preset boundary conditions to obtain static analysis results and dynamic analysis results of the building structure sample; The initial model is trained based on the static analysis result and the dynamic analysis result to obtain the trained response prediction model.

3. The method according to claim 2, characterized in that The sample structure parameters include a geometric structure and a corresponding size, and equipment parameters, wherein the geometric structure is used to indicate a shape structure constituting the building structure, and the corresponding size is used to indicate a size of the shape structure, and the equipment parameters include equipment type, equipment size, and equipment location of equipment set in the building structure sample; Then, a geometric model corresponding to the building structure sample is constructed based on the sample structure parameters of the building structure sample, including: Constructing a three-dimensional model of the building structure sample based on the geometric structure and corresponding dimensions of the building structure sample; The corresponding device model is set in the three-dimensional model based on the device parameters to obtain the geometric model.

4. The method according to claim 2, characterized in that According to the geometric model, the material properties, environmental parameters and preset boundary conditions corresponding to the building structure sample, the static analysis results and dynamic analysis results of the building structure sample are obtained, including: Meshing the geometric model to obtain a finite element model; Calculating the stress, strain and displacement of the preset key structural parts of the building structure sample of the finite element model under static load based on the material properties and the preset boundary conditions to obtain the static analysis results; The vibration response of the preset key structural parts of the building structure sample under the dynamic load of the finite element model is calculated based on the material properties, the environmental parameters and the preset boundary conditions to obtain the dynamic analysis results.

5. The method according to claim 2, characterized in that The initial model includes stacked multi-layer restricted Boltzmann machines; Then, the initial model is trained based on the static analysis result and the dynamic analysis result to obtain the trained response prediction model, including: Using the sample structure parameters of the building structure sample as input layer data, using the static analysis results and the dynamic analysis results as output layer data, training each layer of restricted Boltzmann machine layer by layer based on contrast divergence from the bottom layer to obtain the hierarchical training weights of each layer of restricted Boltzmann machine; wherein the hierarchical training weights of each layer of restricted Boltzmann machine are used as the initialization weights of the next layer of restricted Boltzmann machine; The hierarchical training weight progress is adjusted based on the back propagation algorithm, the target weight of the restricted Boltzmann machine of each layer is determined, and the trained response prediction model is obtained.

6. The method according to claim 5, characterized in that Adjusting the progress of the hierarchical training weights based on the back propagation algorithm to determine the target weight of the restricted Boltzmann machine at each layer includes: w′ is the target weight, w is the layer training weight, η is the learning rate, and loss(y, t) is the loss function between the model prediction value y and the output layer data t.

7. The method according to claim 1, characterized in that The first structural parameter is updated based on the first response and a preset optimization target to obtain a target structural parameter, including: In response to the first response not satisfying the preset optimization target, updating the first structural parameter based on the first response to obtain a second structural parameter; Inputting the second structural parameter into the response prediction model to obtain a corresponding second response; Determining whether the second response meets the preset optimization goal; In response to the second response satisfying the preset optimization target, determining the second structural parameter as the target structural parameter; In response to the second response not satisfying the preset optimization target, the second structural parameter is updated based on the second response.

8. A device for determining structural parameters of a building structure, characterized in that: include: An acquisition module, used for acquiring first structural parameters and preset environmental parameters about the building structure; A prediction module, used for inputting the first structural parameter into a trained response prediction model to obtain a corresponding first response; wherein the response prediction model is trained based on static analysis results and dynamic analysis results of a building structure sample, the static analysis results include a static response of the building structure sample under a static load, and the dynamic analysis results include a vibration response of the building structure sample under a dynamic load; A parameter determination module is used to update the first structural parameters based on the first response and a preset optimization target to obtain target structural parameters.

9. The device according to claim 8, characterized in that Also includes: A model module, used for constructing a geometric model corresponding to the building structure sample based on the sample structure parameters of the building structure sample; An analysis module, used to perform analysis according to the geometric model, material properties corresponding to the building structure sample, environmental parameters and preset boundary conditions, to obtain static analysis results and dynamic analysis results of the building structure sample; The training module is used to train the initial model based on the static analysis results and the dynamic analysis results to obtain the trained response prediction model.

10. The device according to claim 9, characterized in that The sample structure parameters include a geometric structure and a corresponding size, and equipment parameters, wherein the geometric structure is used to indicate a shape structure constituting the building structure, and the corresponding size is used to indicate a size of the shape structure, and the equipment parameters include equipment type, equipment size, and equipment location of equipment set in the building structure sample; Then, a geometric model corresponding to the building structure sample is constructed based on the sample structure parameters of the building structure sample, including: Constructing a three-dimensional model of the building structure sample based on the geometric structure and corresponding dimensions of the building structure sample; The corresponding device model is set in the three-dimensional model based on the device parameters to obtain the geometric model.

11. The device according to claim 9, characterized in that According to the geometric model, the material properties, environmental parameters and preset boundary conditions corresponding to the building structure sample, the static analysis results and dynamic analysis results of the building structure sample are obtained, including: Meshing the geometric model to obtain a finite element model; Calculating the stress, strain and displacement of the preset key structural parts of the building structure sample of the finite element model under static load based on the material properties and the preset boundary conditions to obtain the static analysis results; The vibration response of the preset key structural parts of the building structure sample under the dynamic load of the finite element model is calculated based on the material properties, the environmental parameters and the preset boundary conditions to obtain the dynamic analysis results.

12. The device according to claim 9, characterized in that The initial model includes stacked multi-layer restricted Boltzmann machines; Then, the initial model is trained based on the static analysis result and the dynamic analysis result to obtain the trained response prediction model, including: Using the sample structure parameters of the building structure sample as input layer data, using the static analysis results and the dynamic analysis results as output layer data, training each layer of restricted Boltzmann machine layer by layer based on contrast divergence from the bottom layer to obtain the hierarchical training weights of each layer of restricted Boltzmann machine; wherein the hierarchical training weights of each layer of restricted Boltzmann machine are used as the initialization weights of the next layer of restricted Boltzmann machine; The hierarchical training weight progress is adjusted based on the back propagation algorithm, the target weight of the restricted Boltzmann machine of each layer is determined, and the trained response prediction model is obtained.

13. The device according to claim 12, characterized in that Adjusting the progress of the hierarchical training weights based on the back propagation algorithm to determine the target weight of the restricted Boltzmann machine at each layer includes: w′ is the target weight, w is the layer training weight, η is the learning rate, and loss(y, t) is the loss function between the model prediction value y and the output layer data t.

14. The device according to claim 8, characterized in that The first structural parameter is updated based on the first response and a preset optimization target to obtain a target structural parameter, including: In response to the first response not satisfying the preset optimization target, updating the first structural parameter based on the first response to obtain a second structural parameter; Inputting the second structural parameter into the response prediction model to obtain a corresponding second response; Determining whether the second response meets the preset optimization goal; In response to the second response satisfying the preset optimization target, determining the second structural parameter as the target structural parameter; In response to the second response not satisfying the preset optimization target, the second structural parameter is updated based on the second response.

15. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the program. 16 . A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method according to claim 1 .

Images