TW202211070A - Electronic device and method for accelerating generation of simulation result of simulation software - Google Patents

Abstract

An electronic device and a method for accelerating a generation of a simulation result of a simulation software are provided. The method includes: obtaining input data; generating fake data according to the input data; inputting a first fake feature of the fake data to the simulation software to generate a first fake simulation result; obtaining a user-defined feature; calculating a first similarity between the first fake feature and the user-defined feature; generating, based on the first similarity, an approximate simulation result matching with the user-defined feature according to the first fake simulation result; and outputting the approximate simulation result.

Description

Electronic apparatus and method for accelerating generation of simulation results for simulation software

The present invention relates to an electronic device and method for accelerating the generation of simulation results for simulation software.

Numerical simulation is a combination of theoretical calculations and computer programs, and then a computer is used to perform huge and complex operations. In the process of simulation calculation, it is necessary to use a pair of approximation methods, so that the obtained simulation results can meet the real situation. Therefore, the generation of simulation results must rely on hardware devices with high computing power. As the complexity of the simulation model increases, so does the time required to generate the simulation results. Therefore, simulation software is often unable to provide simulation results to users in real time. Accordingly, how to speed up the generation of simulation results without increasing the hardware construction cost is one of the goals that those in the art are committed to.

The present invention provides an electronic device and method for accelerating the generation of simulation results of simulation software, which can reduce the generation time of the simulation results.

An electronic device for accelerating the generation of simulation results of simulation software of the present invention includes a processor, a storage medium and a transceiver. The transceiver obtains the input data. The storage medium stores multiple modules. The processor is coupled to the storage medium and the transceiver, and accesses and executes a plurality of modules, wherein the plurality of modules includes a machine learning model and a computing module. Machine learning models generate fake data based on input data. The computing module inputs the first imitation feature of the imitation data to the simulation software through the transceiver to generate a first imitation simulation result, receives the user-defined feature through the transceiver, and calculates the first imitation feature and the first imitation between the user-defined feature The similarity, based on the first similarity, generates an approximate simulation result matching the user-defined characteristic according to the first fake simulation result, and outputs the approximate simulation result through the transceiver.

In an embodiment of the present invention, the above-mentioned computing module generates the approximate simulation result according to the first imitation feature, the first imitation simulation result and the user-defined characteristic based on one of interpolation and extrapolation.

In an embodiment of the present invention, the above-mentioned computing module inputs the user-defined feature to the simulation software through the transceiver to generate the final simulation result, and outputs the final simulation result through the transceiver.

In an embodiment of the present invention, the above-mentioned modules further include a database of imitation features and a database of simulation results. The imitation feature database stores a plurality of imitation features including the first imitation feature. The simulation result database stores a plurality of imitation simulation results corresponding to the plurality of imitation features respectively, wherein the computing module calculates the similarity between each of the plurality of imitation features and the user-defined feature, and responds to the first similarity A first fake simulation result corresponding to the first degree of similarity is selected from the plurality of fake simulation results for the highest degree of similarity to generate an approximate simulation result.

In an embodiment of the present invention, the above-mentioned modules further include a history database. The historical database stores historical data, wherein the machine learning model includes a generator and a discriminator, wherein the discriminator is trained according to the historical data, wherein the generator generates candidate fake data according to the input data, and the discriminator determines the candidate fake data as fake data.

In an embodiment of the present invention, the above-mentioned machine learning model is a generative adversarial network.

A method for accelerating the generation of simulation results of simulation software of the present invention includes: obtaining input data; generating fake data according to the input data; inputting a first fake feature of the fake data into the simulation software to generate a first fake simulation result; user-defined features; calculating a first similarity between the first fake features and the user-defined features; generating an approximate simulation result matching the user-defined characteristics according to the first fake simulation results based on the first similarity; and outputting the approximate simulation result.

Based on the above, the present invention can instantly generate approximate simulation results for users' reference before the simulation software takes a long time to generate the simulation results.

FIG. 1 is a schematic diagram of an electronic device 100 for accelerating the generation of simulation results of simulation software according to an embodiment of the present invention. The electronic device 100 may include a processor 110 , a storage medium 120 and a transceiver 130 .

The processor 110 is, for example, a central processing unit (CPU), or other programmable general-purpose or special-purpose micro control unit (micro control unit, MCU), microprocessor (microprocessor), digital signal processing digital signal processor (DSP), programmable controller, application specific integrated circuit (ASIC), graphics processor (graphics processing unit, GPU), image signal processor (image signal processor, ISP) ), image processing unit (IPU), arithmetic logic unit (ALU), complex programmable logic device (CPLD), field programmable gate array (field programmable gate array) , FPGA) or other similar elements or a combination of the above. The processor 110 may be coupled to the storage medium 120 and the transceiver 130 , and access and execute a plurality of modules and various application programs stored in the storage medium 120 .

The storage medium 120 is, for example, any type of fixed or removable random access memory (random access memory, RAM), read-only memory (ROM), and flash memory (flash memory). , a hard disk drive (HDD), a solid state drive (SSD), or similar components or a combination of the above components for storing a plurality of modules or various application programs executable by the processor 110 . In this embodiment, the storage medium 120 can store a plurality of models including an arithmetic module 121 , a machine learning model 122 , a history database 123 , a fake feature database 124 , a simulation result database 125 and a user-defined feature database 126 , etc. group, its function will be explained later.

The transceiver 130 transmits and receives signals in a wireless or wired manner. Transceiver 130 may also perform operations such as low noise amplification, impedance matching, frequency mixing, up or down frequency conversion, filtering, amplification, and the like.

The electronic device 100 can be used to generate fake simulation results. FIG. 2 illustrates a flow chart of generating fake simulation results according to an embodiment of the present invention. In step S210 , the computing module 121 can obtain the input data through the transceiver 130 . Input data can be image data or data curves used to generate simulation results. For example, the input data may include images or curves related to meteorology such as satellite cloud images, radar echo images, temperature curves, humidity curves, rainfall accumulation curves or air quality curves, but the invention is not limited thereto.

In step S220, the machine learning module 122 can generate fake data according to the input data. Fake data can be fake graphics or data curves. For example, the fake data may include images such as a fake satellite image, a fake radar echo image, a fake temperature curve, a fake humidity curve, a fake rainfall accumulation curve, or a fake air quality curve. limited to this. The machine learning model 122 is, for example, a generative adversarial network (GAN). The machine learning model 122 may include a generator (generator) 1221 and a discriminator (discriminator) 1222, wherein the discriminator 122 is trained and trained by the machine learning model 122 according to the historical data of the historical database 123, wherein the historical data includes historical satellite cloud images, Images such as historical radar echo map, historical temperature curve, historical humidity curve, historical rainfall accumulation curve or historical air quality curve.

FIG. 3 is a detailed flow chart illustrating the steps of generating fake data according to input data according to an embodiment of the present invention. In step S221, the generator 1221 may generate candidate fake data according to the input data. In step S222, the discriminator 1222 may compare whether the candidate fake data matches the historical data. If the candidate fake data does not match the historical data, the discriminator 1222 can delete the candidate fake data, and step S221 can be re-executed to generate a new candidate fake data. If the candidate counterfeit data matches the historical data, go to step S223. In step S223 , the discriminator 1222 may determine the candidate fake data as fake data, and store the generated fake data in the fake feature database 124 . Next, step S221 may be re-executed to generate new candidate fake data. Steps S221 to S223 in FIG. 3 can be repeatedly performed until the counterfeit feature database 124 obtains enough counterfeit data.

For example, suppose the input data is a satellite image. The machine learning model 122 can generate a plurality of fake satellite cloud images according to historical data (eg, historical satellite cloud images) and satellite cloud images.

Back to Figure 2. In step S230, the computing module 121 may perform feature extraction on a plurality of imitation data in the imitation feature database 124 to obtain a plurality of imitation features respectively corresponding to the plurality of imitation data. The computing module 121 can store a plurality of imitation features in the imitation feature database 124 .

Specifically, the computing module 122 may perform feature extraction on each of the plurality of simulated satellite cloud images to obtain simulated regional rainfall information (ie, simulated features). The computing module 122 can store the simulated regional rainfall information in the simulated feature database 124 . For example, the pseudo-feature database 124 may store pseudo-area rainfall information corresponding to 400 mm, pseudo-area rainfall information corresponding to 500 mm, and pseudo-area rainfall information corresponding to 600 mm.

In step S240, the computing module 121 may input the plurality of imitation features into the simulation software to generate a plurality of imitation simulation results respectively corresponding to the plurality of imitation features. The computing module 121 can store a plurality of simulated simulation results in the simulation result database 125 .

For example, if the simulation software is used to determine whether a flood occurs, the computing module 122 can respectively input a plurality of simulated regional rainfall information (ie, simulated features) to the simulation software. The simulation software can generate multiple flood judgment results (ie, fake simulation results) corresponding to the rainfall information of multiple fake areas respectively. The imitation feature database 124 can store a plurality of regional rainfall information and a plurality of flood judgment results respectively corresponding to the plurality of regional rainfall information. For example, the fake feature database 124 can store the fake area rainfall information corresponding to 400 mm and the corresponding flood judgment result, the fake area rainfall information corresponding to 500 mm and the corresponding flood judgment result, and the corresponding flood damage judgment result corresponding to 600 mm. Fake the regional rainfall information and the corresponding flood judgment results.

After obtaining a plurality of counterfeit simulation results, the electronic device 100 can generate an approximate simulation result corresponding to the user's requirement according to the plurality of counterfeit simulation results. The generation of approximate simulation results takes only a fraction of the time. In addition, the electronic device 100 can also utilize simulation software to generate a time-consuming but more accurate final simulation result. FIG. 4 illustrates a flow chart for generating approximate simulation results and final simulation results according to an embodiment of the present invention.

In step S410 , the computing module 121 may receive the user-defined characteristics through the transceiver 130 . User-defined characteristics can be determined by the user according to his needs. For example, if the user is interested in the flood judgment result corresponding to the regional rainfall information of 540 mm, the user may decide that the user-defined feature is corresponding to the regional rainfall information of 540 mm. The user may input the user-defined characteristics to the electronic device 100 through an input device (eg, a keyboard or a touch screen). The computing module 121 can obtain the user-defined characteristics from the input device through the transceiver 130 . The computing module 121 may store the user-defined features in the user-defined feature database 126 .

In step S420, the computing module 122 may calculate the similarity between the plurality of imitation features and the user-defined characteristic, and respond to the first similarity between the first imitation feature and the user-defined characteristic among the plurality of imitation features The first imitation feature is selected from the plurality of imitation features as the highest degree of similarity. For example, the computing module 122 may select from the fake regional rainfall information corresponding to 400 mm, the fake regional rainfall information corresponding to 400 mm, The simulated area rainfall information corresponding to 500 mm is selected from the simulated area rainfall information of 500 mm and the simulated area rainfall information corresponding to 600 mm.

In one embodiment, after selecting the first fake feature corresponding to the highest similarity, the computing module 122 may respond that the second similarity between the second fake feature and the user-defined feature among the plurality of fake features is: The second highest similarity is selected from the remaining plurality of imitation features (ie: a plurality of imitation features that have not been selected), and the second imitation feature is selected. For example, after selecting the fake regional rainfall information corresponding to 500 mm, the computing module 122 may respond that the similarity between the fake regional rainfall information corresponding to 600 mm and the regional rainfall information corresponding to 540 mm is the second highest similarity. The simulated area rainfall information corresponding to 600 mm is selected from the simulated area rainfall information corresponding to 400 mm and the simulated area rainfall information corresponding to 600 mm.

In step S430 , the computing module 122 may generate an approximate simulation result matching the user-defined feature according to the first fake simulation result based on the first similarity, and may output the approximate simulation result through the transceiver 130 . Specifically, after selecting the first counterfeit feature according to the first similarity, the computing module 122 may select the first counterfeit simulation result from the plurality of counterfeit simulation results stored in the simulation result database 145 according to the first similarity . Next, the computing module 122 may generate an approximate simulation result according to the first fake feature, the first fake simulation result corresponding to the first fake feature, and the user-defined feature based on the interpolation method or the extrapolation method.

For example, the computing module 122 may determine the flood risk corresponding to the 500 mm fake area rainfall information (ie, the first fake feature) based on the interpolation method or the extrapolation method (ie, the first fake area rainfall information) : the first imitation simulation result) and the regional rainfall information corresponding to 540 mm (ie: user-defined characteristics) to generate the flood judgment result (ie: approximate simulation result) corresponding to the regional rainfall information of 540 mm. Since the approximate simulation result only needs to use the pre-generated first fake simulation result, it is not necessary to use the simulation software. Therefore, approximate simulation results can be generated very quickly.

In one embodiment, after selecting the first imitation feature and the second imitation feature according to the first similarity and the second similarity, the computing module 122 may select the first imitation feature from the simulation result database 145 according to the first similarity. A first fake simulation result is selected from a plurality of fake simulation results, and a second fake simulation result is selected from a plurality of fake simulation results stored in the simulation result database 145 according to the second similarity. Next, the computing module 122 may perform the interpolation method or the extrapolation method according to the first imitation feature, the first imitation simulation result corresponding to the first imitation feature, the second imitation feature, and the second imitation corresponding to the second imitation feature Simulation results and user-defined features produce approximate simulation results.

For example, the computing module 122 may determine the flood risk corresponding to the 500 mm fake area rainfall information (ie, the first fake feature) based on the interpolation method or the extrapolation method (ie, the first fake area rainfall information) : the first fake simulation result), the 600mm fake area rainfall information (ie: the second fake feature), the flood judgment result corresponding to the 600mm fake area rainfall information (ie: the second fake simulation result), and the corresponding 540 mm of regional rainfall information (ie: user-defined characteristics) to generate flood judgment results (ie: approximate simulation results) corresponding to 540 mm of regional rainfall information. Since the approximate simulation result only needs to be generated by using the pre-generated first and second fake simulation results, no simulation software is required. Therefore, approximate simulation results can be generated very quickly.

In order to obtain a final simulation result that is more accurate than the approximate simulation result, in step S440 , the computing module 122 can input the user-defined feature to the simulation software through the transceiver 130 to generate the final simulation result, and can output the final simulation result through the transceiver 130 Simulation results. For example, the computing module 122 can input the regional rainfall information corresponding to 540 mm (ie: user-defined characteristics) to the simulation software through the transceiver 130 to generate a flood judgment result corresponding to the regional rainfall information of 540 mm (ie, : final simulation result).

FIG. 5 is a flowchart illustrating a method for accelerating the generation of simulation results of simulation software according to an embodiment of the present invention, wherein the method may be implemented by the electronic device as shown in FIG. 1 . In step S510, input data is acquired. In step S520, counterfeit data is generated according to the input data. In step S530, the first imitation feature of the imitation data is input to the simulation software to generate a first imitation simulation result. In step S540, user-defined characteristics are acquired. In step S550, a first similarity between the first counterfeit feature and the user-defined feature is calculated. In step S560, an approximate simulation result matching the user-defined feature is generated according to the first fake simulation result based on the first similarity. In step S570, the approximate simulation result is output.

To sum up, the electronic device of the present invention can use the generative adversarial network to generate a plurality of counterfeit features, and input the plurality of counterfeit features to the simulation software to generate a plurality of counterfeit simulation results in advance. When the user wants to obtain a simulation result for a specific feature, the user can input the user-defined feature to the electronic device. The electronic device may compare the similarity between the user-defined feature and a plurality of pre-stored counterfeit features, and select one or more counterfeit features that are highly correlated with the user-defined feature. The electronic device may calculate approximate simulation results according to one or more counterfeit simulation results corresponding to one or more counterfeit features. Since the approximate simulation results do not need to be generated by the simulation software, the approximate simulation results are generated very quickly. In addition to generating approximate simulation results, the electronic device can also input user-defined features into simulation software to generate more accurate simulation results.

100: Electronics 110: Processor 120: Storage Media 121: Operation module 122: Machine Learning Models 1221: Generator 1222: Discriminator 123: Historical Database 124: Counterfeit feature database 125: Simulation Results Database 126: User-defined feature database 130: Transceiver S210, S220, S221, S222, S223, S230, S240, S410, S420, S430, S440, S510, S520, S530, S540, S550, S560, S570: Steps

FIG. 1 is a schematic diagram of an electronic device for accelerating the generation of simulation results of simulation software according to an embodiment of the present invention. FIG. 2 illustrates a flow chart of generating fake simulation results according to an embodiment of the present invention. FIG. 3 is a detailed flow chart illustrating the steps of generating fake data according to input data according to an embodiment of the present invention. FIG. 4 illustrates a flow chart for generating approximate simulation results and final simulation results according to an embodiment of the present invention. FIG. 5 is a flowchart illustrating a method for accelerating the generation of simulation results of simulation software according to an embodiment of the present invention.

S510, S520, S530, S540, S550, S560, S570: Steps

Claims (7)

An electronic device for accelerating the generation of simulation results for simulation software, comprising: Transceiver to obtain input data; storage media, storing multiple modules; and a processor, coupled to the storage medium and the transceiver, and accessing and executing the multiple modules, wherein the multiple modules include: a machine learning model that generates fake data based on the input data; and an arithmetic module, for inputting a first imitation feature of the imitation data to the simulation software through the transceiver to generate a first imitation simulation result, receiving a user-defined feature through the transceiver, and calculating the first imitation a first degree of similarity between the feature and the user-defined characteristic, generating an approximate simulation result matching the user-defined characteristic from the first fake simulation result based on the first degree of similarity, and by the The transceiver outputs the approximate simulation result. The electronic device of claim 1, wherein the computing module is based on one of an interpolation method and an extrapolation method according to the first counterfeit feature, the first counterfeit simulation result, and the user definition Features produce the approximate simulation results. The electronic device of claim 1, wherein the computing module inputs the user-defined feature to the simulation software through the transceiver to generate a final simulation result, and outputs the final simulation result through the transceiver Simulation results. The electronic device of claim 1, wherein the modules further comprise: A counterfeit feature database, storing a plurality of counterfeit features including the first counterfeit feature; A simulation result database storing a plurality of simulation results corresponding to the plurality of counterfeit features, wherein the computing module calculates the similarity between each of the plurality of counterfeit features and the user-defined feature degree, and in response to the first degree of similarity being the highest degree of similarity, the first fake simulation result corresponding to the first degree of similarity is selected from the plurality of fake simulation results to generate the approximate simulation result. The electronic device of claim 1, wherein the modules further comprise: Historical database, which stores historical data, among which The machine learning model includes a generator and a discriminator, wherein the discriminator is trained based on the historical data, wherein the generator generates candidate fake data based on the input data, wherein the discriminator determines the candidate Counterfeit material is said counterfeit material. The electronic device of claim 5, wherein the machine learning model is a generative adversarial network. A method of accelerating the generation of simulation results for a simulation software body, comprising: obtain input data; generating fake data based on said input data; inputting the first counterfeit feature of the counterfeit data to the simulation software to generate a first counterfeit simulation result; obtain user-defined characteristics; calculating a first similarity between the first fake feature and the user-defined feature; generating an approximate simulation result that matches the user-defined characteristic based on the first fake simulation result based on the first similarity; and The approximate simulation result is output.

Images