A simulation model display method and device

Abstract

The present application discloses a simulation model display method and device, which is applied to the field of simulation technology. In the present application, target model data of a three-dimensional simulation model is sent to a browser via a cloud platform. If a user's sectioning control operation on the three-dimensional simulation model is detected by the browser, during the sectioning control operation, the first section model data corresponding to the section currently cut out of the three-dimensional simulation model is determined based on the sectioning position information corresponding to the sectioning control operation at the current moment; at least part of the data in the first section model data is subjected to dimensionality reduction processing, and the second section model data obtained by the dimensionality reduction processing is sent to the browser, so that the browser can display the sectioned three-dimensional simulation model by loading the second section model data. The solution of the present application can display the sectioned simulation model in a timely manner during the user's sectioning operation on the simulation model, thereby reducing the occurrence of freezes in the display screen of the simulation model.

Description

Technical Field

[0001] The present application relates to the field of simulation technology, and in particular to a simulation model display method and device. Background Art

[0002] Computer-aided engineering (CAE) and other simulation technologies can be used to simulate objects such as equipment and machine parts. Because object simulation requires large amounts of data and the hardware resources of terminal devices are relatively limited, the use of cloud platforms to implement object simulation is becoming increasingly widespread.

[0003] After the cloud platform outputs the simulation model of the object to the browser, the user can perform a sectioning operation on the displayed simulation model through the browser to understand the internal structure and characteristics of the object through the sectioning surface of the simulation model. Taking into account the limitations of network bandwidth, the cloud platform will generally transmit the model data related to the sectioned simulation model to the browser after confirming that the sectioning control operation is completed, so that the browser can load the relevant data and process it, and finally display the sectioned simulation model. However, transmitting the data related to the sectioned simulation model only after confirming that the sectioning operation is complete will inevitably result in the browser being unable to present the sectioned simulation model in a timely manner. In particular, when the user performs multiple sectioning operations on the simulation model through continuous sectioning operations, the simulation model after each sectioning cannot be presented. Summary of the Invention

[0004] In view of the above problems, the present application provides a simulation model display method and device, so that when a user performs a sectioning operation on a simulation model, the sectioned simulation model can be displayed in a timely manner.

[0005] The first aspect of the present application provides a simulation model display method, which is applied to a cloud platform, comprising:

[0006] Sending target model data of the three-dimensional simulation model to the browser, wherein the target model data is used by the browser to display the three-dimensional simulation model;

[0007] If a sectioning control operation of the three-dimensional simulation model by the user is detected by the browser, during the sectioning control operation, first section model data corresponding to a section currently cut out of the three-dimensional simulation model is determined based on sectioning position information corresponding to the sectioning control operation at the current moment;

[0008] performing dimensionality reduction processing on at least a portion of the first section model data to obtain second section model data;

[0009] Send the second cross-section model data to the browser so that the browser displays the sliced three-dimensional simulation model based on the second cross-section model data.

[0010] In a possible case of the first aspect, the first cross-section model data includes: the basic composition data of the cross-section plane and at least one attribute feature data;

[0011] The dimension reduction processing of at least part of the data in the first cross-section model data to obtain the second cross-section model data includes:

[0012] Perform dimension reduction processing on each type of attribute feature data in the first cross-section model data respectively to obtain the second cross-section model data, and the second cross-section model data includes the basic composition data of the cross-section plane and at least one type of attribute feature data after dimension reduction processing.

[0013] In another possible case of the first aspect, the dimension reduction processing of at least part of the data in the first cross-section model data to obtain the second cross-section model data includes:

[0014] Use an encoder to perform dimension reduction processing on at least part of the data in the first cross-section model data to obtain the second cross-section model data;

[0015] Wherein, the encoder is a neural network model trained according to model data samples of at least one three-dimensional simulation model sample according to a set training target, and the training target is that the decoding result corresponding to the model data sample after dimension reduction processing by the encoder has the smallest gap with the model data sample.

[0016] In another possible case of the first aspect, the encoder is trained in the following manner:

[0017] Obtain model data samples of at least one three-dimensional simulation model sample;

[0018] For each model data sample, use the encoder to perform dimension reduction processing on the model data sample to obtain the dimension-reduced model data sample;

[0019] Use a decoder to decode the dimension-reduced model data sample to obtain the decoded predicted model data sample;

[0020] Based on each model data sample and the predicted model data sample corresponding to each model data sample, adjust the model parameters of the encoder and the decoder, and return to execute the operation of using the encoder to perform dimension reduction processing on the model data sample until the training target is met, and obtain the trained encoder.

[0021] In yet another possible scenario of the first aspect, determining the first cross-sectional model data corresponding to the cross-sectional plane currently cut out from the three-dimensional simulation model based on the cross-sectional position information corresponding to the current cross-section control operation includes:

[0022] Obtaining the cross-sectional position information corresponding to the cross-section control operation sent by the browser based on a set sampling frequency;

[0023] Based on the cross-sectional position information corresponding to the cross-section control operation obtained at the current moment and the target model data, determining the first cross-sectional model data corresponding to the cross-sectional plane currently cut out from the three-dimensional simulation model.

[0024] A second aspect of the present application provides a simulation model display method, which is applied to a browser and includes:

[0025] Displaying the three-dimensional simulation model based on the target model data of the three-dimensional simulation model sent by the cloud platform;

[0026] If it is detected that the user performs a cross-section control operation on the three-dimensional simulation model, during the operation process of the cross-section control operation, obtaining the cross-sectional position information corresponding to the current moment of the cross-section control operation according to a set sampling frequency and sending it to the cloud platform;

[0027] Obtaining the second cross-sectional model data returned by the cloud platform, where the second cross-sectional model data is obtained by the cloud platform performing dimensionality reduction processing on at least part of the data in the first cross-sectional model data, and the first cross-sectional model data is the model data corresponding to the cross-sectional plane currently cut out from the three-dimensional simulation model determined by the cloud platform based on the cross-sectional position information of the cross-section control operation;

[0028] Based on the second cross-sectional model data, displaying the three-dimensional simulation model after cross-sectioning.

[0029] In a possible scenario of the second aspect, it further includes:

[0030] When it is detected that the cross-section control operation stops or ends, performing dimensionality increase processing on the second cross-sectional model data obtained most recently to obtain the third cross-sectional model data;

[0031] Based on the third cross-sectional model data, displaying the three-dimensional simulation model after cross-sectioning.

[0032] In yet another possible scenario of the second aspect, the second cross-sectional model data is obtained by using an encoder to perform dimensionality reduction processing on at least part of the data in the first cross-sectional model data;

[0033] The performing dimensionality increase processing on the second cross-sectional model data obtained most recently includes:

[0034] Using a decoder to perform dimension-upgrading processing on the second cross-section model data obtained most recently;

[0035] In which, the decoder and the encoder are model data samples based on at least one three-dimensional simulation model sample, and are trained according to the set training objectives. The training objectives are the decoding results corresponding to the model data samples after dimensionality reduction processing by the encoder, and the gap with the model data samples is the smallest. The decoding results are the data obtained by the decoder by dimensionality increase processing of the model data samples after dimensionality reduction processing.

[0036] A third aspect of the present application provides a simulation model display device, applied to a cloud platform, comprising:

[0037] A model output unit, configured to send target model data of a three-dimensional simulation model to a browser, wherein the target model data is used by the browser to display the three-dimensional simulation model;

[0038] a data determining unit configured to, if a section control operation of the user on the three-dimensional simulation model is detected through the browser, determine, during the section control operation, first section model data corresponding to a section plane currently sectioned from the three-dimensional simulation model based on section position information corresponding to the section control operation at a current moment;

[0039] a dimensionality reduction processing unit, configured to perform dimensionality reduction processing on at least a portion of the first section model data to obtain second section model data;

[0040] The data sending unit is used to send the second section model data to the browser, so that the browser displays the cut three-dimensional simulation model based on the second section model data.

[0041] A fourth aspect of the present application provides another simulation model display device, applied to a browser, comprising:

[0042] a first display unit, configured to display the three-dimensional simulation model based on target model data of the three-dimensional simulation model sent by the cloud platform;

[0043] a data sending unit configured to, upon detecting a user's section control operation on the three-dimensional simulation model, obtain section position information corresponding to the section control operation at the current moment according to a set sampling frequency during the section control operation and send the information to the cloud platform;

[0044] a data obtaining unit, configured to obtain second section model data returned by the cloud platform, the second section model data being obtained by the cloud platform performing dimensionality reduction processing on at least a portion of the first section model data, the first section model data being model data corresponding to a section plane currently sectioned in the three-dimensional simulation model, determined by the cloud platform based on section position information of the section control operation;

[0045] The second display unit is used to display the cut three-dimensional simulation model based on the second section model data.

[0046] From the above content, it can be seen that when the user performs a sectioning control operation on the three-dimensional simulation model through the browser, the cloud platform determines the first sectioning model data corresponding to the sectioning surface of the three-dimensional simulation model currently being sectioned based on the sectioning position information of the sectioning control operation at the current moment, and then performs dimensionality reduction processing on the first sectioning model data before transmitting it to the browser. This can not only reduce the amount of data transmission between the cloud platform and the browser and reduce the data transmission delay; it can also reduce the amount of data that the browser needs to process, so that the sectioned three-dimensional simulation model can be presented in a timely manner during the sectioning control operation, and the display screen of the three-dimensional simulation model can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the originals and elements are not necessarily drawn to scale.

[0048] Figure 1 A schematic diagram of a process of applying the simulation model display method provided in this application to the cloud platform side;

[0049] Figure 2 A schematic diagram of a process of applying the simulation model display method provided in this application to the browser side;

[0050] Figure 3 A schematic diagram of a process interaction of the simulation model display method provided in this application;

[0051] Figure 4 This is a comparison example diagram of the point cloud of the three-dimensional simulation model in this application before and after dimensionality reduction processing;

[0052] Figure 5 This is a comparative example diagram of the cloud image of the three-dimensional simulation model in this application before and after dimensionality reduction processing;

[0053] Figure 6A schematic diagram of the structure of a simulation model display device provided in this application;

[0054] Figure 7 A schematic diagram of the structure of another simulation model display device provided by this application;

[0055] Figure 8 A schematic diagram of the structure of the electronic device provided in this application. DETAILED DESCRIPTION

[0056] The embodiments of the present application are described below in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation methods of the present application are only used to explain the specific embodiments of the present application and are not intended to limit the present application. It is known to those skilled in the art that with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0057] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0058] In this application, a cloud platform is used to process data calculations related to 3D simulation models. For example, simulation software can be deployed on the cloud platform. Based on the simulation software components, the cloud platform can not only assist in building a 3D simulation model of an object, but also simulate the sectioning of the 3D simulation model and calculate data related to the sectioning.

[0059] On this basis, users can log in to the cloud platform through a browser and obtain the simulation services provided by the cloud platform. By leveraging the powerful computing resources of the cloud platform, they can efficiently calculate relevant data during the cutting process of the three-dimensional simulation model, thus realizing the Software as a Service (SAAS) service model.

[0060] The following first introduces the simulation model display method of this application from the cloud platform side.

[0061] like Figure 1 , shows a flow chart of a simulation model display method provided by an embodiment of the present application. The method of this embodiment is applied to a cloud platform, and this embodiment may include:

[0062] S101, sending target model data of a three-dimensional simulation model to a browser.

[0063] The target model data is used by the browser to display the three-dimensional simulation model. Based on this, the browser can display the three-dimensional simulation model by loading the target model data.

[0064] The target model data may include various data required to present the three-dimensional simulation model.

[0065] For example, the target model data may include: basic component data of a three-dimensional simulation model, which may include the identification of the grid nodes included in the three-dimensional simulation model and the position information of each grid node, and may also include geometric shape information, stress distribution, strain distribution and other information of the three-dimensional simulation model, without limitation.

[0066] The target model data may also include at least one attribute feature information of the three-dimensional simulation model. Each attribute feature information may include an attribute feature associated with each grid node in the three-dimensional simulation model. For example, the at least one attribute feature information may include but is not limited to: attribute feature information such as temperature, velocity, density, and energy associated with each grid node in the three-dimensional simulation model.

[0067] In this application, the target model data of the 3D simulation model can be uploaded to the cloud platform by a user via a browser; the 3D simulation model can also be a 3D simulation model built by a user on the cloud platform using a browser. There is no restriction on the specific implementation of how the cloud platform obtains the target model data of the 3D simulation model.

[0068] It is understandable that, before step S101 , the cloud platform may also establish a communication connection with the client where the browser is located, and the specific process of establishing the communication connection is not limited.

[0069] S102, if the browser detects the user's cutting control operation on the three-dimensional simulation model, during the cutting control operation, based on the cutting position information corresponding to the cutting control operation at the current moment, the first section model data corresponding to the cutting surface currently cut out of the three-dimensional simulation model is determined.

[0070] The sectioning control operation may be input by a user through a browser and is used to control the sectioning of the three-dimensional simulation model.

[0071] For example, the sectioning control operation can be an operation of controlling a sectioning tool (such as a simulated section plane or other forms of tools, which is not limited thereto) in the simulation display window through a mouse or a button. Herein, the simulation display window is a window for displaying a three-dimensional simulation model. By controlling the sectioning tool to pass through a specific part of the three-dimensional simulation model, the three-dimensional simulation model can be sectioned.

[0072] For another example, the sectioning control operation can also be an operation of controlling the sectioning tool to section the three-dimensional simulation model in response to a user inputting parameters in a simulation control interface displayed by a browser. Herein, the sectioning method, number of times, etc. of the sectioning tool on the three-dimensional simulation model are related to the input parameters set by the user. Based on this, if the sectioning operation of the sectioning tool does not stop, it can be considered that the sectioning control operation is still in the operation process.

[0073] It can be understood that the user can interact with the cloud platform through the input sectioning control operation to achieve the purpose of viewing the section plane of the three-dimensional simulation model. In practical applications, the user may need to view a section of the three-dimensional simulation model through the sectioning control operation, or may need to continuously view multiple sections of the three-dimensional simulation model through the sectioning control operation. Based on this, according to different requirements of the user for viewing the internal structure of the three-dimensional simulation model, the sectioning control operation on the three-dimensional simulation model can be to section the three-dimensional simulation model only once, or may be to continuously section multiple times.

[0074] It can be understood that since the sectioning control operation is a continuous operation behavior that may last for a certain period of time, during the operation process of the sectioning control operation, the sectioning position information corresponding to the sectioning control operation will also change continuously. In order to enable the browser side to timely display the sectioned three-dimensional simulation model, the cloud platform will, during the operation process of the sectioning control operation, determine in real time the first section model data corresponding to the section plane sectioned from the three-dimensional simulation model based on the sectioning position information corresponding to the sectioning control operation.

[0075] For example, during the operation process of the sectioning control operation, the browser will continuously collect the sectioning position information corresponding to the sectioning control operation at the current moment and send it to the cloud platform. And each time the cloud platform obtains the sectioning position information corresponding to the current moment of the sectioning control operation sent by the browser, it will execute an operation of determining the first section model data.

[0076] In one possible implementation, during the operation of the cutting control operation, the browser can obtain the cutting position information corresponding to the cutting control operation at the current moment according to the set sampling frequency and send it to the cloud platform. The set sampling frequency is the sampling frequency at which the browser collects relevant data parameters in the simulation control window, and is also the frequency at which the browser reports data to the cloud platform. Accordingly, the cloud platform can obtain the cutting position information corresponding to the cutting control operation sent by the browser based on the set sampling frequency. On this basis, the cloud platform can determine the first section model data corresponding to the cutting plane currently cut out of the three-dimensional simulation model based on the cutting position information corresponding to the cutting control operation obtained at the current moment and the target model data.

[0077] The set sampling frequency may be pre-specified by the cloud platform. Of course, it may also be determined in other ways, which is not limited.

[0078] The cutting position information corresponding to the cutting control operation is used to represent the cutting position at which the three-dimensional simulation model is cut by the cutting control operation.

[0079] For example, the slicing position information may include coordinate position information and normal information corresponding to the slicing control operation, i.e., coordinate position information and normal information of the slicing control operation within the coordinate system corresponding to the 3D simulation model. Based on this, the coordinate position information and normal information corresponding to the slicing control operation can represent the intersection position information of the slicing control operation and the 3D simulation model. This intersection position information can reflect the position of the slicing plane created by slicing the 3D simulation model.

[0080] Of course, the sectioning position information may also be the relative position information between the sectioning control operation determined by the browser and the 3D simulation model, or other information that can represent the sectioning position of the 3D simulation model, and there is no limitation to this.

[0081] It can be understood that during the operation of the cutting control operation, the three-dimensional simulation model may be cut multiple times, and the first section model data is the model data corresponding to the cutting surface generated after the three-dimensional simulation model is cut at the current moment. Based on the first section model data, the current cutting surface can be presented.

[0082] The first cross-sectional model data contains the same data types as the target model data. For example, the first cross-sectional model data may include basic component data corresponding to the cross-sectional surface and at least one type of attribute characteristic data. The meanings of the basic component data and attribute characteristic information are the same as those described above and are not further elaborated.

[0083] In this application, there is no limitation on the specific implementation method for determining the first cross-sectional model data. For example, based on the cross-sectional position information of the cross-section control operation at the current moment, the first cross-sectional model data corresponding to the cross-section obtained by cutting the current 3D simulation model can be extracted from the target model data.

[0084] It can be understood that a 3D simulation model is constructed by multiple meshes, and a mesh is composed of mesh nodes, and each mesh node is located at a vertex of the mesh. On this basis, based on the cross-sectional position information of the cross-section control operation at the current moment, the cross-section of the 3D simulation model at the current moment can be determined, and thus the mesh nodes forming the cross-section can be determined. Then, based on the target model data, the data associated with each mesh node belonging to the cross-section can be determined, and the first cross-sectional model data can be constructed based on the data associated with each mesh node.

[0085] S103. Perform dimensionality reduction processing on at least part of the data in the first cross-sectional model data to obtain second cross-sectional model data.

[0086] Among them, the second cross-sectional model data is the data obtained by performing dimensionality reduction processing on at least part of the data in the first cross-sectional model data. Therefore, the second cross-sectional model data includes part of the data in the first cross-sectional model data that has undergone dimensionality reduction processing and another part of the data that has not undergone dimensionality reduction processing.

[0087] In this application, the purpose of performing dimensionality reduction processing on at least part of the data in the first cross-sectional model data is to convert at least part of the data in the first cross-sectional model data from high-dimensional data to low-dimensional data and try to retain the important information in the first cross-sectional model data.

[0088] It can be understood that by performing dimensionality reduction processing on at least part of the data in the first cross-sectional model data, the data volume of the first cross-sectional model data can be reduced. Therefore, the data volume of the second cross-sectional model data is smaller than that of the first cross-sectional model data.

[0089] In this application, dimensionality reduction processing can be performed on all or part of the first cross-sectional model data according to actual needs.

[0090] In a possible implementation manner, considering that the data volume of the basic composition data forming the 3D simulation model (or the cross-section) is relatively small and has a relatively large impact on the overall shape structure of the 3D simulation model (or the cross-section), while the data volume of the attribute feature data is relatively large. Therefore, in this application, dimensionality reduction processing can be performed on each type of attribute feature data in the first cross-sectional model data respectively to obtain the second cross-sectional model data. Correspondingly, the second cross-sectional model data includes the basic composition data in the first cross-sectional model data and at least one type of attribute feature data after dimensionality reduction processing.

[0091] For example, each type of attribute feature information corresponding to the cross-section includes the attribute feature information of each grid node on the cross-section. Then, for each type of attribute feature information, dimensionality reduction processing can be performed on the attribute feature information of each grid node in the cross-section, and then the dimensionality-reduced attribute feature information is re-assigned to each grid node on the cross-section. Finally, the model data after dimensionality reduction processing corresponding to the cross-section, that is, the second cross-section model data, is obtained.

[0092] In this application, there can be multiple possible specific implementation manners for performing dimensionality reduction processing on at least part of the data in the first cross-section model data, and no limitation is imposed thereon. In one possible implementation manner, in order to reduce the data volume of the first cross-section model data and be able to retain the data characteristics of the first cross-section model data to the greatest extent, an encoder can also be pre-trained in this application. On this basis, the encoder can be used to perform dimensionality reduction processing on at least part of the data in the first cross-section model data to obtain the second cross-section model data.

[0093] Among them, the encoder can be a neural network model trained according to the model data samples of at least one three-dimensional simulation model sample according to a set training objective. The training objective is that the decoding result corresponding to the model data sample after dimensionality reduction processing by the encoder has the smallest difference from the model data sample.

[0094] Among them, the decoding result corresponding to the model data sample after dimensionality reduction processing is the decoded model data sample obtained by decoding the model data sample after dimensionality reduction processing.

[0095] It can be understood that the result of performing dimensionality reduction processing on the model data sample by the encoder is actually the encoding result of the encoder for encoding the model data sample. On this basis, in order to be able to restore the model data sample before dimensionality reduction based on the model data sample after dimensionality reduction processing, it is necessary to decode the model data sample after dimensionality reduction processing.

[0096] In this application, a neural network model is trained as an encoder, and the training objective for training the encoder is that the decoding result corresponding to the model data sample after dimensionality reduction processing by the encoder has the smallest difference from the model data sample. Therefore, the trained encoder can not only perform compression encoding on at least part of the data in the first cross-section model data, but also enable the second cross-section model data obtained by compression encoding to retain the main characteristics of the first cross-section model data to the greatest extent.

[0097] S104. Send the second cross-section model data to the browser so that the browser can display the sliced three-dimensional simulation model based on the second cross-section model data.

[0098] As previously explained, while the second cross-sectional model data only has a relatively lower dimensionality than the first cross-sectional model data, the second cross-sectional model data still includes the primary data features of the first cross-sectional model data. Therefore, based on the second cross-sectional model data, the browser can present the primary features of the cross-sectional surface of the three-dimensional simulation model after sectioning. Furthermore, unlike conventional methods of compressing data into compressed data packets, in this application, the browser does not need to decompress the second cross-sectional model data, and can directly load the second cross-sectional model data to present the three-dimensional simulation model after sectioning.

[0099] The inventors of this application have discovered that when a user slices a 3D simulation model using a slicing control, the 3D simulation model constantly changes with the slicing control operation, and therefore, the user does not pay attention to the detailed features of the 3D simulation model after the slice. On this basis, because the second sectional model data retains the main data features of the first sectional model data, the browser can present the main features of the sectional surface of the 3D simulation model after the slice based on the second sectional model data, allowing the user to still see the important features of the 3D simulation model after the slice, without affecting the user's viewing experience.

[0100] Moreover, since the data volume of the second section model data is smaller than that of the first section model data, the cloud platform takes less time to transmit the second section model data to the browser, and when the browser loads and processes the second section model data, the amount of data required to be processed is smaller, which can effectively improve the display efficiency of the three-dimensional simulation model after cutting and reduce the display delay of the three-dimensional simulation model.

[0101] It should be noted that during the operation of the cutting control operation, the cloud platform will execute steps S103 and S104 each time it determines the first section model data, so that the second section model data that needs to be sent to the browser can be continuously updated based on the cutting position information of the cutting control operation, so that the browser side can promptly display the three-dimensional simulation model after the current cutting.

[0102] From the above content, it can be seen that when the user performs sectioning control operations on the three-dimensional simulation model through the browser, the cloud platform determines the first sectioning model data corresponding to the sectioning surface of the three-dimensional simulation model currently being sectioned based on the sectioning position information of the sectioning control operation at the current moment, and then performs dimensionality reduction processing on the first sectioning model data before transmitting it to the browser. This not only reduces the amount of data transmission between the cloud platform and the browser and reduces data transmission delay; it also reduces the amount of data that the browser needs to process, thereby enabling the timely presentation of the sectioned three-dimensional simulation model during the sectioning control operation and reducing the occurrence of freezes in the display of the three-dimensional simulation model.

[0103] In addition, during the section control operation, the cloud platform only transmits to the browser the model data after dimension reduction corresponding to the currently sectioned section plane. The amount of data transmitted is small, so that when the user inputs a section control operation by sliding or other means, the sectioned section planes can be presented smoothly, reducing the situation of stuttering in the display of the section planes.

[0104] The following introduces the solution of the present application from the browser side. As Figure 2 , it shows another flow diagram of the simulation model display method provided by the embodiment of the present application. The method of this embodiment can be applied to a browser, and the browser can also be called a browser client. This embodiment may include:

[0105] S201, based on the target model data of the three-dimensional simulation model sent by the cloud platform, display the three-dimensional simulation model.

[0106] S202, if it is detected that the user performs a section control operation on the three-dimensional simulation model, during the operation of the section control operation, obtain the section position information corresponding to the current moment of the section control operation according to the set sampling frequency and send it to the cloud platform.

[0107] Among them, the section position information can refer to the relevant introduction in the previous cloud platform side embodiment and will not be elaborated here.

[0108] It can be understood that when the browser detects a section control operation, it can report the event of detecting the section control operation to the cloud platform or notify the cloud platform in other ways, so that the cloud platform can confirm that the browser has detected the section control operation. At the same time, during the operation of the section control operation, the browser will collect the section position information corresponding to the section control operation according to the set sampling frequency and report it, so that the cloud platform can obtain the change of the section position information in time during the operation of the section control operation, thereby continuously updating the first section model data corresponding to the section plane of the currently sectioned three-dimensional simulation model, and returning the currently determined first section model data to the browser.

[0109] S203, obtain the second section model data returned by the cloud platform.

[0110] Among them, the second section model data is obtained by the cloud platform performing dimension reduction processing on at least part of the data in the first section model data. The first section model data is the model data corresponding to the section plane currently sectioned from the three-dimensional simulation model determined by the cloud platform based on the section position information of the section control operation.

[0111] Among them, the specific implementation of the cloud platform for determining the second section model data can refer to the relevant introduction in the previous cloud platform side and will not be elaborated here.

[0112] S204. Based on the second cross-section model data, display the three-dimensional simulation model after sectioning.

[0113] For example, based on the sectioning position information of each sectioning control operation, the browser can determine the remaining part model of the three-dimensional simulation model after sectioning. This remaining part model does not include the current sectioning plane. On this basis, the browser determines the model data corresponding to this remaining part model from the target model data, loads the model data corresponding to this remaining part model and the second cross-section model data, and performs rendering processing to present the three-dimensional simulation model with this sectioning plane after sectioning.

[0114] Among them, the amount of calculation for the browser to determine the model data of the remaining part model after sectioning the three-dimensional simulation model is relatively small and the time consumption is less. However, the complexity of determining the model data corresponding to the sectioning plane is relatively high and the amount of calculation is relatively large. Therefore, by having the cloud platform determine the first cross-section model data of this sectioning plane and perform dimensionality reduction processing, the powerful computing power of the cloud platform can be utilized to reduce the time consumption.

[0115] Of course, there may be other possibilities for the browser to present the three-dimensional simulation model after sectioning based on this second cross-section model data, and there is no limit to this.

[0116] From the above content, it can be seen that after the browser side detects the sectioning control operation on the three-dimensional simulation model, it only needs to obtain the sectioning position information corresponding to the current moment of the sectioning control operation at the set sampling frequency during the operation process of the sectioning control operation and send it to the cloud platform, and then it can obtain the second cross-section model data corresponding to the sectioning plane generated by the current sectioning of the three-dimensional simulation model returned by the cloud platform. Thus, the powerful data computing power of the cloud platform can be utilized to efficiently determine the model data corresponding to the sectioning plane of the three-dimensional simulation model and reduce the computing time consumption required to determine the model data of the sectioning plane.

[0117] Moreover, since the second cross-section model data transmitted by the cloud platform is data after dimensionality reduction and the data volume is relatively small, it can not only reduce the time consumption required for data transmission, but also reduce the data processing volume of the browser. Naturally, it can reduce the time consumption required for the browser to display the three-dimensional simulation model after sectioning, so that the browser can more timely display the three-dimensional simulation model after sectioning and reduce the situation of screen freezing during the process of performing the sectioning operation on the three-dimensional simulation model.

[0118] It is understood that when a user pauses or ends sectioning a 3D simulation model, the user may wish to view various detailed features, such as the attributes, of the sectioned 3D simulation model. Based on this, upon detecting that the sectioning control operation has stopped (i.e., paused) or ended, the browser may further perform dimensionality-upgrading on the most recently acquired second section model data to obtain third section model data. Accordingly, the browser may display the sectioned 3D simulation model based on this third section model data.

[0119] The dimensionality upgrade process for the second section model data is to increase the data dimension of the second section model data so that the third section model data obtained by the dimensional upgrade can contain richer feature information of the three-dimensional model after sectioning.

[0120] It is understandable that because the second and third cross-sectional model data are different, the three-dimensional simulation model obtained after sectioning based on the second cross-sectional model data is different from the three-dimensional simulation model obtained after sectioning based on the third cross-sectional model data. Because the third cross-sectional model data has a higher data dimension and contains richer data features, the detailed information of the cross-sectional surface on the three-dimensional simulation model after sectioning based on the third cross-sectional model data is richer.

[0121] For example, during the sectioning control operation, the cloud platform performs dimensionality reduction processing on the attribute feature data, such as temperature and density, of each mesh node in the 3D simulation model. Therefore, the browser displays relatively few attribute feature details of the sectioned 3D simulation model based on the second section model data. However, by increasing the dimensionality of the attribute feature data in the third section model data obtained by dimensionality upgrading the second section model data, the attribute feature details of the section surface in the sectioned 3D simulation model obtained based on the third section model data are more detailed.

[0122] It is understandable that during the operation of the section control operation, the cloud platform will continuously obtain the second section model data sent by the browser, and when the section control operation stops or ends, the user needs to see the specific details of the three-dimensional simulation model obtained by the most recent section. Therefore, it is necessary to use the most recently obtained second section model data to determine the third section model data.

[0123] In the present application, there are many possibilities for the specific implementation of dimensionality upgrading of the second cross-sectional model data. The simulation model display method of the present application is introduced below in combination with one implementation method.

[0124] like Figure 3 , shows a schematic diagram of a process interaction of a simulation model display method provided by an embodiment of the present application. The method of this embodiment may include:

[0125] S301, the cloud platform sends target model data of the three-dimensional simulation model to the browser.

[0126] S302: The browser displays the three-dimensional simulation model based on the target model data.

[0127] S303, if the browser detects the user's sectioning control operation on the three-dimensional simulation model, during the sectioning control operation, the browser obtains the sectioning position information corresponding to the sectioning control operation at the current moment according to the set sampling frequency and sends it to the cloud platform.

[0128] S304, during the operation of the cutting control operation, the cloud platform determines the first section model data corresponding to the cutting plane currently cut out of the three-dimensional simulation model based on the cutting position information corresponding to the cutting control operation at the current moment.

[0129] The first section model data includes: basic composition data of the section cut out at the current moment and at least one attribute feature data.

[0130] S305: The cloud platform uses an encoder to perform dimensionality reduction processing on each attribute feature data in the first section model data to obtain second section model data, and sends the second section model data to the browser.

[0131] For example, for each attribute feature data , the dimension-reduced data converted by the encoder It can be expressed as follows:

[0132] ;

[0133] in, is the function representation of the encoder, are the parameters of the encoder.

[0134] S306: The browser displays the cut 3D simulation model based on the second section model data.

[0135] Among them, using an encoder to perform dimensionality reduction processing on various attribute feature information in the first section model data can reduce the data volume of the first section model data, so that the data volume of the obtained second section model data is smaller than the data volume of the first section model data. This can not only reduce the data transmission volume and data transmission delay; it can also reduce the time required for the browser to load model data, thereby greatly reducing the delay in displaying the three-dimensional simulation model after cutting.

[0136] In addition, in the present application, by training the encoder, during the process of reducing the dimensionality of each attribute feature data in the first cross-sectional model data using the encoder, the main model features can be retained, so that the sliced three-dimensional simulation model can still present the main features when displayed.

[0137] For ease of understanding, taking the example that the browser side needs to present the density of each grid node in the three-dimensional simulation model as an attribute feature, correspondingly, the browser can load the basic data of the sliced three-dimensional simulation model and the density feature data after dimensionality reduction processing. On this basis, reference can be made to Figure 4 and Figure 5 , in Figure 4 and Figure 5 the darkness of the color represents the density size, the darker the color, the greater the density.

[0138] Among them, Figure 4 shows a comparison example diagram of the point cloud of the three-dimensional simulation model before and after dimensionality reduction processing by the encoder. In Figure 4 , the left point cloud image is the point cloud before dimensionality reduction processing, while the right is the point cloud image after dimensionality reduction processing. It can be seen from the comparison that before and after dimensionality reduction processing, the basic structure of the point cloud remains unchanged, and the density features of the point cloud are also similar.

[0139] Similarly, Figure 5 shows a comparison example diagram of the cloud map of the three-dimensional simulation model before and after dimensionality reduction processing by the encoder. The left cloud map is the cloud map before dimensionality reduction processing corresponding to the three-dimensional simulation model, while the right cloud map is the cloud map after dimensionality reduction processing. It can be seen that the density features of the cloud map are similar before and after dimensionality reduction processing.

[0140] S307. When it is detected that the slicing control operation stops or ends, the browser uses the decoder to perform dimensionality increase processing on the second cross-sectional model data obtained most recently to obtain the third cross-sectional model data.

[0141] Among them, the decoder and the encoder are obtained by training according to the model data samples of at least one three-dimensional simulation model sample according to the set training objective. As described above, the training objective is that the decoding result corresponding to the model data sample after dimensionality reduction processing by the encoder has the smallest gap with the model data sample. In this embodiment, the decoding result is the data obtained by the decoder performing dimensionality increase processing on the model data sample after dimensionality reduction processing.

[0142] In the present application, the encoder and the decoder can be trained synchronously, and the specific implementation manner of training the encoder and the decoder can be not limited.

[0143] For example, in a possible implementation manner, the encoder and the decoder can be obtained by training in the following manner:

[0144] First, obtain model data samples of at least one three-dimensional simulation model sample;

[0145] Secondly, for each model data sample, use an encoder to perform dimensionality reduction processing on the model data sample to obtain the dimensionality-reduced model data sample;

[0146] Then, use a decoder to decode the dimensionality-reduced model data sample to obtain the decoded predicted model data sample;

[0147] Finally, based on each model data sample and the predicted model data sample corresponding to each model data sample, adjust the model parameters of the encoder and decoder, and return to perform the operation of using the encoder to perform dimensionality reduction processing on the model data sample until the set training objective is met, and obtain the trained encoder and decoder.

[0148] Among them, the three-dimensional simulation model sample is a three-dimensional simulation model used as a training sample. Correspondingly, the model data of the three-dimensional simulation model sample is called the model data sample.

[0149] It can be understood that using the encoder to perform dimensionality reduction processing on the model data sample is also using the encoder to encode the model data sample. Therefore, in order to restore the model data sample before dimensionality reduction, it is necessary to use the decoder to decode the dimensionality-reduced model data sample. And the essence of decoding is to perform dimensionality increase processing on the dimensionality-reduced model data sample.

[0150] Among them, the training objective is as described above.

[0151] There can be multiple possibilities for the specific implementation of determining whether the training objective is met. For example, when the number of training iterations reaches the set number, or the loss function value determined based on each model data sample and the predicted model data sample corresponding to each model data sample converges, it can be determined that the training objective is met.

[0152] Among them, there can be multiple possibilities for the loss function for determining the loss function value, and there is no restriction on this. For example, taking the model data sample including the original data samples of each grid node in the three-dimensional simulation model sample as an example, it can be understood that if only the attribute feature data is encoded, the original data sample can be regarded as a kind of attribute feature data. On this basis, the mean square error function can be selected as the loss function , specifically expressed as follows:

[0153]

[0154] Among them, represents the original data sample of the th grid node in the three-dimensional simulation model sample, the The data obtained by encoding the original data samples of a grid node and then decoding them through a decoder. is the number of grid nodes in the three-dimensional simulation model sample.

[0155] S308, the browser displays the sliced three-dimensional simulation model based on the third cross-section model data.

[0156] This step can be referred to the previous introduction and will not be elaborated here.

[0157] Corresponding to the operations on the cloud platform side of the simulation model display method of the present application, the present application also provides a simulation model display device. As Figure 6 , shows a schematic structural diagram of a composition of the simulation model display device provided by an embodiment of the present application. This device can be applied to a cloud platform and includes:

[0158] A model output unit 601, configured to send target model data of a three-dimensional simulation model to the browser, where the target model data is used for the browser to display the three-dimensional simulation model;

[0159] A data determination unit 602, configured to, if it is detected through the browser that a user performs a slicing control operation on the three-dimensional simulation model, during the operation of the slicing control operation, determine first cross-section model data corresponding to the sliced cross-section of the three-dimensional simulation model based on the slicing position information corresponding to the current moment of the slicing control operation;

[0160] A dimensionality reduction processing unit 603, configured to perform dimensionality reduction processing on at least part of the data in the first cross-section model data to obtain second cross-section model data;

[0161] A data sending unit 604, configured to send the second cross-section model data to the browser, so that the browser displays the sliced three-dimensional simulation model based on the second cross-section model data.

[0162] In a possible implementation manner, the first cross-section model data determined by this data determination unit includes: the basic composition data of the cross-section and at least one attribute feature data;

[0163] This dimensionality reduction processing unit includes:

[0164] A dimensionality reduction processing subunit, configured to perform dimensionality reduction processing on each type of attribute feature data in the first cross-section model data respectively to obtain second cross-section model data, where the second cross-section model data includes the basic composition data of the cross-section and at least one type of attribute feature data after dimensionality reduction processing.

[0165] In yet another possible implementation, the dimensionality reduction processing unit is specifically configured to use an encoder to perform dimensionality reduction processing on at least part of the data in the first profile model data to obtain second profile model data;

[0166] Wherein, the encoder is a neural network model trained according to model data samples of at least one 3D simulation model sample according to a set training objective, and the training objective is that the decoding result corresponding to the model data sample after dimensionality reduction processing by the encoder has the smallest gap from the model data sample.

[0167] In yet another possible implementation, the present application further includes: a training unit, configured to train the encoder in the following manner:

[0168] Obtain model data samples of at least one 3D simulation model sample;

[0169] For each model data sample, use the encoder to perform dimensionality reduction processing on the model data sample to obtain a dimensionality-reduced model data sample;

[0170] Use a decoder to decode the dimensionality-reduced model data sample to obtain a decoded predicted model data sample;

[0171] Based on each model data sample and the predicted model data sample corresponding to each model data sample, adjust the model parameters of the encoder and the decoder, and return to execute the operation of using the encoder to perform dimensionality reduction processing on the model data sample until the training objective is met, and obtain the trained encoder.

[0172] In yet another possible implementation, the data determination unit includes:

[0173] A data acquisition subunit, configured to acquire the section position information corresponding to the section control operation sent by the browser based on a set sampling frequency;

[0174] A data extraction subunit, configured to determine the first profile model data corresponding to the section plane currently sectioned out of the 3D simulation model based on the section position information corresponding to the section control operation obtained at the current moment and the target model data.

[0175] On the other hand, corresponding to the operation of the simulation model display method of the present application on the browser side, the present application further provides yet another simulation model display device. As Figure 7 shows another schematic structural diagram of a composition of a simulation model display device provided by the present application. The device of this embodiment can be applied to a browser and includes:

[0176] A first display unit 701, configured to display the 3D simulation model based on the target model data of the 3D simulation model sent by the cloud platform;

[0177] The data sending unit 702 is configured to, upon detecting a user's section control operation on the three-dimensional simulation model, obtain section position information corresponding to the section control operation at the current moment according to a set sampling frequency during the section control operation and send the information to the cloud platform;

[0178] A data obtaining unit 703 is configured to obtain second section model data returned by the cloud platform, where the second section model data is obtained by the cloud platform performing dimensionality reduction processing on at least a portion of the first section model data, where the first section model data is model data corresponding to a section plane currently cut into the three-dimensional simulation model, determined by the cloud platform based on the section position information of the section control operation;

[0179] The second display unit 704 is configured to display the sectioned three-dimensional simulation model based on the second section model data.

[0180] In one possible implementation, the Figure 7 The simulation model display device shown also includes:

[0181] a dimension-increasing processing unit configured to, upon detecting that the sectioning control operation stops or ends, perform dimension-increasing processing on the second section model data obtained most recently to obtain third section model data;

[0182] The third display unit is used to display the cut three-dimensional simulation model based on the third section model data.

[0183] In another possible implementation, the second section model data is obtained by performing dimensionality reduction processing on at least part of the first section model data using an encoder;

[0184] The dimensionality-increasing processing unit includes:

[0185] a dimensionality-increasing processing subunit, configured to, upon detecting that the section control operation stops or ends, perform dimensionality-increasing processing on the second section model data obtained most recently using a decoder;

[0186] Among them, the decoder and encoder are model data samples based on at least one three-dimensional simulation model sample, and are trained according to the set training objectives. The training objective is the decoding result corresponding to the model data sample after dimensionality reduction processing by the encoder, and the difference with the model data sample is the smallest. The decoding result is the data obtained by the decoder by performing dimensionality increase processing on the model data sample after dimensionality reduction processing.

[0187] The embodiment of the present application also provides an electronic device, which can be a device node in the cloud platform or a terminal device where the browser is located.Figure 8 As shown, it shows a schematic diagram of a composition structure of the electronic device, which at least includes a processor 801 and a memory 802;

[0188] The processor 801 is used to execute the simulation model display method described in any one of the above embodiments;

[0189] The memory 802 is used to store the programs required for the processor to perform operations.

[0190] It can be understood that the electronic device may further include a display unit 803 and an input unit 804.

[0191] Of course, the electronic device may also have Figure 8 more or fewer components, and there is no limitation on this.

[0192] An embodiment of the present application also provides a computer program product, including computer-readable instructions. When the computer-readable instructions run on an electronic device, the electronic device is enabled to implement any one of the simulation model display methods provided in the embodiments of the present application.

[0193] An embodiment of the present application also provides a computer-readable storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can be enabled to implement any one of the simulation model display methods provided in the embodiments of the present application.

[0194] In addition, it should be noted that the device embodiments described above are only illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided in the present application, the connection relationship between the modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines.

[0195] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus necessary general hardware. Of course, it can also be implemented by dedicated hardware including application-specific integrated circuits, dedicated CPUs, dedicated memories, dedicated components, etc. Generally, functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structures for implementing the same function can also be various, such as analog circuits, digital circuits, or dedicated circuits. However, for the present application, software program implementation is a better embodiment in more cases. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a readable storage medium, such as a floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disc of a computer, and includes several instructions to enable a computer device (which can be a personal computer, training device, or network device, etc.) to execute the methods described in various embodiments of the present application.

[0196] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product.

[0197] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general computer, a dedicated computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website, computer, training device, or data center to another website, computer, training device, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can store, or a data storage device such as a training device or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state drive (SSD)).

Claims

1. A simulation model display method, characterized in that Applied to cloud platforms, including: Sending target model data of the three-dimensional simulation model to the browser, wherein the target model data is used by the browser to display the three-dimensional simulation model; If a sectioning control operation of the three-dimensional simulation model by the user is detected through the browser, during the sectioning control operation, first section model data corresponding to a section currently cut out of the three-dimensional simulation model is determined based on sectioning position information corresponding to the sectioning control operation at the current moment, the first section model data including: basic composition data of the sectioning surface and at least one attribute feature data, the attribute feature data including attribute features associated with each mesh node on the sectioning surface, and the sectioning control operation is used to perform at least one continuous sectioning of the three-dimensional simulation model; performing dimensionality reduction processing on at least a portion of the first section model data to obtain second section model data; Sending the second section model data to the browser, so that the browser displays the sectioned three-dimensional simulation model based on the second section model data; The step of performing dimensionality reduction processing on at least a portion of the first section model data to obtain the second section model data includes: Dimensionality reduction processing is performed on each attribute feature data in the first section model data to obtain second section model data, which includes basic component data of the section surface and at least one attribute feature data after dimensionality reduction processing.

2. The simulation model display method according to claim 1, characterized in that The step of performing dimensionality reduction processing on each attribute feature data in the first section model data to obtain second section model data includes: Using an encoder to perform dimensionality reduction processing on each type of attribute feature data in the first section model data to obtain second section model data; In which, the encoder is a neural network model obtained by training according to a set training objective based on a model data sample of at least one three-dimensional simulation model sample, and the training objective is that the decoding result corresponding to the model data sample after dimensionality reduction processing by the encoder has the smallest difference with the model data sample.

3. The simulation model display method according to claim 2, characterized in that, The encoder is trained in the following way: Obtaining a model data sample of at least one three-dimensional simulation model sample; For each model data sample, using an encoder to perform dimensionality reduction processing on the model data sample to obtain a model data sample after dimensionality reduction; Decoding the reduced-dimensional model data sample using a decoder to obtain a decoded prediction model data sample; Based on each model data sample and the predicted model data sample corresponding to each model data sample, the model parameters of the encoder and the decoder are adjusted, and the operation of using the encoder to perform dimensionality reduction processing on the model data sample is returned to execute until the training objective is met, thereby obtaining a trained encoder.

4. The simulation model display method according to claim 1, characterized in that The determining, based on the cutting position information corresponding to the cutting control operation at the current moment, first section model data corresponding to the cutting plane currently cut out of the three-dimensional simulation model includes: Obtaining the slicing position information corresponding to the slicing control operation sent by the browser based on a set sampling frequency; Based on the sectioning position information corresponding to the current sectioning control operation and the target model data obtained at the current moment, determine the first section model data corresponding to the section plane currently sectioned out from the three-dimensional simulation model.

5. A simulation model display method, characterized in that, Applied to a browser, it includes: Based on the target model data of the three-dimensional simulation model sent by the cloud platform, display the three-dimensional simulation model; If it is detected that the user performs a sectioning control operation on the three-dimensional simulation model, during the operation process of the sectioning control operation, obtain the sectioning position information corresponding to the current moment of the sectioning control operation at a set sampling frequency and send it to the cloud platform, where the sectioning control operation is used to perform at least one continuous sectioning on the three-dimensional simulation model; Obtain the second section model data returned by the cloud platform, where the second section model data is obtained by the cloud platform performing dimensionality reduction processing on various attribute feature data in the first section model data. The first section model data is the model data corresponding to the section plane currently sectioned out from the three-dimensional simulation model determined by the cloud platform based on the sectioning position information of the sectioning control operation. The first section model data includes: the basic composition data of the section plane and at least one type of attribute feature data, and the attribute feature data includes the attribute features associated with each mesh node on the section plane; the second section model data includes the basic composition data of the section plane and at least one type of attribute feature data after dimensionality reduction processing; Based on the second section model data, display the sectioned three-dimensional simulation model.

6. The simulation model display method according to claim 5, characterized in that It further includes: When it is detected that the sectioning control operation stops or ends, perform dimensionality increase processing on the second section model data obtained most recently to obtain the third section model data; Based on the third section model data, display the sectioned three-dimensional simulation model.

7. The simulation model display method according to claim 6, wherein The second section model data is obtained by using an encoder to perform dimensionality reduction processing on various attribute feature data in the first section model data; The performing dimensionality increase processing on the second section model data obtained most recently includes: Using a decoder to perform dimensionality increase processing on the second section model data obtained most recently; Wherein, the decoder and the encoder are trained based on the model data samples of at least one three-dimensional simulation model sample according to a set training objective. The training objective is that the decoding result corresponding to the model data sample after dimensionality reduction processing by the encoder has the smallest gap with the model data sample, and the decoding result is the data obtained by the decoder performing dimensionality increase processing on the dimensionality-reduced model data sample.

8. A simulation model display device, characterized in that, Applied to a cloud platform, it includes: A model output unit, configured to send the target model data of the three-dimensional simulation model to the browser, where the target model data is used for the browser to display the three-dimensional simulation model; a data determining unit configured to, if a user's sectioning control operation on the three-dimensional simulation model is detected through the browser, determine, during the sectioning control operation, first section model data corresponding to a section currently cut out of the three-dimensional simulation model based on sectioning position information corresponding to the sectioning control operation at a current moment, wherein the first section model data includes: basic composition data of the sectioning surface and at least one attribute feature data, the attribute feature data including attribute features associated with each mesh node on the sectioning surface, wherein the sectioning control operation is used to perform at least one continuous sectioning of the three-dimensional simulation model; a dimensionality reduction processing unit, configured to perform dimensionality reduction processing on at least a portion of the first section model data to obtain second section model data; a data sending unit, configured to send the second section model data to the browser, so that the browser displays the sectioned three-dimensional simulation model based on the second section model data; Wherein, the dimensionality reduction processing unit includes: The dimensionality reduction processing subunit is used to perform dimensionality reduction processing on each attribute feature data in the first section model data to obtain second section model data, which includes basic component data of the section surface and at least one attribute feature data after dimensionality reduction processing.

9. A simulation model display device, characterized in that, Applies to browsers, including: a first display unit, configured to display the three-dimensional simulation model based on target model data of the three-dimensional simulation model sent by the cloud platform; a data sending unit configured to, upon detecting a user's sectioning control operation on the three-dimensional simulation model, obtain sectioning position information corresponding to the sectioning control operation at the current moment according to a set sampling frequency during the sectioning control operation, and send the information to the cloud platform, wherein the sectioning control operation is used to perform at least one continuous sectioning of the three-dimensional simulation model; a data acquisition unit, configured to obtain second section model data returned by the cloud platform, the second section model data being obtained by the cloud platform through dimensionality reduction processing of various attribute feature data in the first section model data, the first section model data being model data corresponding to a section plane currently cut out of the three-dimensional simulation model, determined by the cloud platform based on section position information of the section control operation, the first section model data comprising: basic component data of the section plane and at least one attribute feature data, the attribute feature data comprising attribute features associated with each grid node on the section plane; the second section model data comprising the basic component data of the section plane and at least one attribute feature data after dimensionality reduction processing; The second display unit is used to display the cut three-dimensional simulation model based on the second section model data.

Images