Methods, systems, and computer programs (generating digital simulation scenarios using sampling)

Abstract

To improve performance of a computer simulation using sampling.SOLUTION: The present invention includes processing, using a computer, a resolution reduction of input data by an upscaling technique providing upscaling of the input data. The input data includes a size and a resolution, a location, a date, conditioning points, and weather variable data. A multiple-point geostatistics technique is applied to reduced resolution data to generate a simulation of a new weather scenario without changes in size or resolution to the reduced resolution data. The computer is used to process a resolution increase of the reduced resolution data from the simulation using a downscaling technique to return the reduced resolution data to the size and the resolution of the input data.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to systems and methods for generating computerized weather scenarios using sampling techniques.

[0002] Computational systems and computer-implemented methods can be used to generate weather scenarios using sampled data. However, even with the availability of hardware capable of multiple calculations, it may be infeasible to reduce the computational cost for large amounts of data. One problem that can arise when simulating weather scenarios over large areas is that the simulation may produce very large matrices, for example, more than 250 x 250 data points. Summary of the Invention [Problem to be solved by the invention]

[0003] The present disclosure recognizes the shortcomings and problems associated with current techniques for producing computer-generated weather simulations covering large areas that create data with large matrices. [Means for solving the problem]

[0004] The present invention alleviates the problem of generating computer simulation scenarios covering large areas, which creates data with very large matrices, exceeding 250 x 250 points. The invention solves the problem of the high computational cost of implementing multi-point geostatistical methods for weather generation scenarios, without focusing on approaches involving hardware availability. The present invention uses a system or method for generating computerized weather scenarios using optimal multi-point geostatistical methods.

[0005] In an aspect according to the invention, a method for improving the performance of a computer simulation using sampling includes: processing, using a computer, a reduced resolution of input data using an upscaling technique to provide upscaling of the input data, the input data including size and resolution, location, date, conditioning point, and meteorological variable data; the upscaling of the input data generates reduced-resolution data having a reduced resolution and reduced data size that is lower resolution and smaller data size than the input data; applying a multi-point geostatistical technique to the reduced-resolution data to generate, using the computer, a simulation of a new meteorological scenario without modifying the reduced-resolution data by the reduced data size or the reduced resolution; and processing, using the computer, a resolution increase of the reduced-resolution data from the simulation using a downscaling technique to return the reduced-resolution data to the size and resolution of the input data.

[0006] In a related aspect, the simulation is a digital weather simulation based on the reduced resolution data.

[0007] In a related aspect, the input data includes location information including latitude and longitude, conditioning points, and weather-related information including weather information including weather conditions and dates.

[0008] In a related aspect, the weather variables and conditions are received from a user's computer and are associated with a particular location.

[0009] In a related aspect, the generated simulation is the same size as the size of the input data as received at the computer.

[0010] In a related aspect, the multipoint geostatistical technique is direct sampling or parallel direct sampling.

[0011] In a related aspect, the upscaling is performed using a method of representing one point based on a set of points.

[0012] In a related aspect, the downscaling is performed using a method to represent a set of points based on a neighborhood or based on a previous set of data associated with the location.

[0013] In a related aspect, the method further comprises, if there is a database and a pre-trained neural network available for the requested data, using the neural network to perform the upscaling or downscaling; and if the neural network is not available, the upscaling is performed using bicubic resampling or average pooling, and the downscaling is performed using an interpolation-based method using one data sample.

[0014] In a related aspect, the method further comprises performing upsampling and downsampling using one or more methods other than a generative neural network if the neural network is unavailable; and performing upsampling and downsampling using the generative neural network if the neural network is available with training data.

[0015] In another aspect according to the invention, a system for improving the performance of a computer simulation using sampling includes a computer system having a computer processor, a computer-readable storage medium, and program instructions, the program instructions being stored on the computer-readable storage medium and executable by the processor to cause the computer system to perform the following functions: using a computer to process resolution reduction of input data using an upscaling technique to provide upscaling of the input data, the input data including data size and resolution, location, date, conditioning point, and meteorological variable data; the upscaling of the input data to generate reduced-resolution data having a reduced resolution and reduced data size that is lower resolution and smaller data size than the input data; applying a multi-point geostatistical technique to the reduced-resolution data to generate simulations of new meteorological scenarios without changing the reduced data size or the reduced resolution of the reduced-resolution data; and using a computer to process resolution increase of the reduced-resolution data from the simulation using a downscaling technique to return the reduced-resolution data to the size and resolution of the input data.

[0016] In a related aspect, the simulation is a digital weather simulation based on the reduced resolution data.

[0017] In a related aspect, the input data is weather-related information including location information, including latitude and longitude, conditioning points, and weather information, including weather conditions and dates.

[0018] In a related aspect, the weather variables and the conditions are received from a user's computer and are associated with a particular location.

[0019] In a related aspect, the generated simulation is the same size as the size of the input data as received at the computer.

[0020] In a related aspect, the multipoint geostatistical technique is direct sampling or parallel direct sampling.

[0021] In a related aspect, the upscaling is performed using a method of representing one point based on a set of points.

[0022] In a related aspect, the downscaling is performed using a method to represent a set of points based on a neighborhood or based on a previous set of data associated with the location.

[0023] In a related aspect, the system further comprises steps for using a neural network to perform the upscaling or downscaling if a database and a pre-trained neural network are available for the requested data; and steps for performing the upscaling using bicubic resampling or average pooling and the downscaling using an interpolation-based method using one data sample if the neural network is not available.

[0024] In another aspect according to the present invention, a computer program product for improving performance of a computer simulation using sampling includes a computer-readable storage medium having program instructions embodied thereon. The program instructions are executable by the computer to cause the computer to perform computer-implemented functions including a step for computer-processing resolution reduction of input data using an upscaling technique to provide upscaling of the input data. The input data includes size and resolution, location, date, conditioning point, and meteorological variable data, and the upscaling of the input data generates reduced-resolution data having a lower resolution and reduced data size than the input data. The computer program product includes a step for computer-generating simulations of new meteorological scenarios without modifying the reduced-resolution data by applying a multi-point geostatistical technique to the reduced-resolution data. The computer program product includes a step for computer-processing resolution increase of the reduced-resolution data from the simulation using a downscaling technique to restore the reduced-resolution data to the size and resolution of the input data. [Brief explanation of the drawings]

[0025] These and other objects, features, and advantages of the present invention will become apparent from the following detailed description of illustrative embodiments thereof, which is to be read in conjunction with the accompanying drawings. Various features of the drawings are not to scale, as the illustrations are for clarity in order to facilitate understanding of the invention by those skilled in the art in conjunction with the detailed description. The drawings are discussed below.

[0026] [Figure 1]1 is a flowchart of a process for generating a digital simulation of a weather scenario using sampling techniques, according to one embodiment of the present disclosure.

[0027] [Figure 2] FIG. 1 is a schematic block diagram illustrating a system for generating a digital simulation of a weather scenario using sampling techniques, according to one embodiment of the present disclosure.

[0028] [Figure 3] FIG. 1 is a schematic block diagram illustrating a system for preparing a neural network according to one embodiment of the present disclosure.

[0029] [Figure 4] FIG. 1 is a schematic block diagram illustrating a system for generating a digital simulation of a weather scenario using sampling techniques, according to one embodiment of the present disclosure.

[0030] [Figure 5] 5 is a flowchart of a method according to one embodiment of the present disclosure that can use the system shown in FIG. 4 for generating a digital simulation of a weather scenario using sampling techniques.

[0031] [Figure 6] FIG. 1 is a schematic block diagram illustrating a computer system according to one embodiment of the present disclosure that includes cloud computing components and functionality and that can cooperate with the systems and methods shown in the figures and described herein. DETAILED DESCRIPTION OF THE INVENTION

[0032] In one embodiment according to the present disclosure, a method for improving the performance of a computer simulation using sampling includes computationally processing reduced-resolution input data using an upscaling technique to provide upscaling of the input data. The input data includes size and resolution, location, date, conditioning point, and meteorological variable data. Upscaling the input data generates reduced-resolution data having a lower resolution and a smaller data size than the input data. The method includes computationally applying a multiple-point geostatistics technique to the reduced-resolution data to generate a simulation of a new meteorological scenario without changing the reduced data size or reduced resolution of the reduced-resolution data. The method includes computationally processing increased resolution of the reduced-resolution data from the simulation using a downscaling technique to restore the reduced-resolution data to the size and resolution of the input data, thereby improving the performance of the computer simulation using sampling.

[0033] In a related aspect, the simulation is a digital weather simulation based on reduced resolution data, whereby the method uses the reduced resolution data as detailed in the method.

[0034] In a related aspect, the input data includes weather-related information including location information, including latitude and longitude, conditioning points, and weather information, including weather conditions and dates, thereby further defining the input data.

[0035] In a related aspect, the weather variables and conditions are received from a user's computer and are associated with a particular location, whereby the data is further described and, in one example, received from a user's computer.

[0036] In a related feature, the generated simulation is the same size as the size of the input data as it was received at the computer, thereby further defining the generated simulation.

[0037] In a related aspect, the multipoint geostatistical technique is a direct sampling method or a parallel direct sampling method, thereby providing an example of a multipoint geostatistical technique.

[0038] In a related aspect, the upscaling is performed using a method for representing one point based on a set of points, thereby providing an example for performing the upscaling.

[0039] In a related aspect, downscaling is performed using a method to represent a set of points based on a neighborhood or based on a previous set of data related to the location, thereby presenting one example for performing downscaling.

[0040] In a related feature, the method further includes performing the upscaling or downscaling using a database and pre-trained neural network available for the requested data, if one exists; and if a neural network is not available, the upscaling is performed using bicubic resampling or average pooling, and the downscaling is performed using an interpolation-based method using one data sample, thereby presenting alternative techniques for upscaling and downscaling.

[0041] In a related feature, the method further includes performing upsampling and downsampling using one or more methods other than a generative neural network if a neural network is unavailable; and performing upsampling and downsampling using a generative neural network if a neural network with training data is available, thereby presenting alternative techniques for upsampling and downsampling.

[0042] In another embodiment according to the present disclosure, a system for improving the performance of a computer simulation using sampling includes a computer system having a computer processor, a computer-readable storage medium, and program instructions, the program instructions being stored on the computer-readable storage medium and executable by the processor to cause the computer system to perform the following functions: a computer-based procedure for processing reduced resolution of input data using an upscaling technique to provide upscaling of the input data, the input data including size and resolution, location, date, conditioning point, and meteorological variable data; upscaling the input data to generate reduced-resolution data having a lower resolution and smaller data size than the input data; applying multi-point geostatistical techniques to the reduced-resolution data to generate simulations of new meteorological scenarios without applying any reduced data size or resolution changes to the reduced-resolution data; and a computer-based procedure for processing increased resolution of reduced-resolution data from the simulation using a downscaling technique to restore the reduced-resolution data to the size and resolution of the input data, thereby improving the performance of the computer simulation using sampling.

[0043] In a related aspect, the simulation is a digital weather simulation based on reduced resolution data, whereby the system uses the reduced resolution data as detailed above.

[0044] In a related aspect, the input data is weather-related information including location information, including latitude and longitude, conditioning points, and weather information, including weather conditions and dates, thereby further defining the input data.

[0045] In a related aspect, the weather variables and conditions are received from the user's computer and are associated with a particular location, thereby further describing the input data received from the user's computer.

[0046] In a related feature, the generated simulation is the same size as the size of the input data as it was received at the computer, thereby further defining the generated simulation.

[0047] In a related aspect, the multipoint geostatistical technique is a direct sampling method or a parallel direct sampling method, thereby providing an example of a multipoint geostatistical technique.

[0048] In a related aspect, the upscaling is performed using a method for representing a point based on a set of points, thereby further defining an example of upscaling.

[0049] In a related aspect, downscaling is performed using a method to represent a set of points based on a neighborhood or based on a previous set of data related to the location, thereby providing an example of downscaling.

[0050] In a related aspect, the system further includes: if a database and a pre-trained neural network is available for the requested data, using the neural network to perform upscaling or downscaling; and if a neural network is not available, upscaling is performed using bicubic resampling or average pooling, and downscaling is performed using an interpolation-based method using one data sample, thereby presenting an example of using neural networks or alternative means for upscaling and downscaling.

[0051] In another embodiment according to the present disclosure, a computer program product for improving performance of a computer simulation using sampling includes a computer-readable storage medium having program instructions embodied thereon. The program instructions are executable by a computer to cause the computer to perform computer-implemented functions, including a function for computer-implemented processing of resolution reduction of input data using an upscaling technique to provide upscaling of the input data. The input data includes size and resolution, location, date, conditioning point, and meteorological variable data, and upscaling the input data generates reduced-resolution data having a lower resolution and smaller data size than the input data. The computer program product includes a procedure for computer-implemented processing of the resolution increase of the reduced-resolution data from the simulation using a downscaling technique to restore the reduced-resolution data to the size and resolution of the input data, thereby improving performance of the computer simulation using sampling. Embodiments and Examples

[0052] The following description, which refers to the accompanying drawings, is provided to aid in a comprehensive understanding of exemplary embodiments of the present invention, as defined by the claims and their equivalents. While the description herein includes numerous specific details to aid in such understanding, these details should be considered merely as examples and are intended to provide clarity and conciseness. Accordingly, those skilled in the art will recognize that various changes and modifications to the embodiments described herein can be made without departing from the scope and spirit of the present invention. Furthermore, descriptions of commonly known functions and configurations may be omitted.

[0053] The terms and phrases used in the following description and claims are not limited to their bibliographical meanings, but are used merely to enable a clear and consistent understanding of the present invention. Therefore, it should be apparent to those skilled in the art that the following description of exemplary embodiments of the present invention is provided for illustrative purposes only, and not for the purpose of limiting the present invention, which is defined by the appended claims and their equivalents.

[0054] The singular forms "a," "an," and "the" should be understood to include plural referents unless the context clearly dictates otherwise. So, for example, reference to a "component surface" includes a reference to one or more of such surfaces unless the context clearly dictates otherwise.

[0055] The embodiments and figures of the present disclosure may have the same or similar components as other embodiments. Such figures and descriptions illustrate and describe further examples and embodiments according to the present disclosure. The embodiments of the present disclosure may include operational actions and / or procedures. A method, such as a computer-implemented method, may include a series of operational blocks for implementing an embodiment according to the present disclosure, which may include cooperation with one or more systems shown in the figures. The operational blocks of methods and systems according to the present disclosure may include techniques, mechanisms, modules, and the like for implementing the functionality of the operations according to the present disclosure. Similar components may have the same reference numerals. Components may operate in cooperation with a computer-implemented method. It is understood that a customer may be an individual, a group of individuals, or a business or organization.

[0056] According to one embodiment of the present disclosure, method 100 can improve the performance of the direct sampling algorithm shown in FIG. 1 for generating weather scenarios at computation time by providing a system for reducing scale. Method 100 can acquire data from user 102, e.g., a user using a computer, in operation 104. The data can include location data, climate variables, or condition points. Method 100 optimizes its use for weather scenario generation by reducing input data size using upscaling techniques in operation 106, applies direct sampling as an embodiment of a multi-point geostatistical method to create a new scenario, as in operation 108, and then performs downscaling using an appropriate method to return to the initial size, i.e., the input data size, as in operation 110. The new data resulting from downscaling can be transmitted to user 102, as shown in FIG. 1. The new data is a new scenario that can be associated with the same latitude and longitude as the input data and with the same resolution as the input data. The new data is the result of downscaling (increasing resolution) as the input data has been subjected to upscaling (decreasing resolution), so that the new data has the same resolution as the input data.

[0057] More specifically, with regard to conditioning points, conditioning points can be used to modify scenarios by setting climate variable values ​​for specific locations. For example, conditioning point A can be represented as [03 / 02 / 2018, [-45, 112], 40 mm], indicating that on date 3-2-2018, at location [-45, 112], there should be 40 mm of precipitation. Such conditions can be communicated to the system to generate new scenarios, where a user can communicate or input data, which may include climate variables, dates, and locations.

[0058] More specifically, with respect to climate variables, the climate variables may include precipitation data, wind data, humidity data, temperature data, and the like.

[0059] In one example, scale relates to resolution, whereby how many physical measurements, e.g., meters, fit within one pixel. For example, a satellite image sign may have a resolution of 30 m, so one pixel represents 30 meters. Size refers to how many pixels the image has, e.g., 256 pixels x 256 pixels. In another example, a user can insert input data, e.g., 256 pixels x 256 pixels, and obtain a new scenario from the system that has 256 pixels x 256 pixels.

[0060] For further clarification, upsampling is understood to be equivalent to downscaling, which results in an increase in the resolution of the data. Upscaling is equivalent to downsampling, which results in a decrease in the resolution of the data.

[0061] It is further understood that upsampling and downsampling can be simple or complex, where simple methods may include average pooling (e.g., calculating the average of several windows [3x3] to obtain the lowest resolution), or interpolation (e.g., extrapolating several additional points between two points to increase resolution). Complex methods may include super-resolution neural network models with many co-located, past, low-resolution, and high-resolution samples to be trained.

[0062] The disclosed method thereby avoids the use and processing of large amounts of data by scaling up and down computer simulations, thereby saving computational bandwidth, data space, and processing resources.

[0063] In one embodiment, a system according to the present disclosure can provide a user with available upscaling / downscaling methods and can request a desired location for creating a new weather scenario. The system will determine whether a database and neural network exists for the requested data, and if so, perform downscaling using the neural network; if not, downscaling is performed using a method that requires only one data sample as input data, as shown in FIG. 3.

[0064] In one embodiment according to the present disclosure, a system or method for generating computerized weather scenarios using an optimal multi-point geostatistical method provides a client API (Application Programming Interface) that returns new weather scenarios depending on the location, date, weather variables, and conditions communicated to the system or method by a user, for example, using a control computer. The system or method can perform upsampling and downsampling, allowing for rapid direct sampling. The system and method provide a system for performing downsampling using a generative neural network. The system or method provides a system that can replace direct sampling methods with parallel methods if there are available cores. The system and method offer the possibility to perform upsampling and downsampling using methods such as average pooling and interpolation, or more complex methods using a generative neural network that require training data.

[0065] Referring to FIG. 2 , in accordance with one embodiment of the present disclosure, a system 199 generates weather scenarios using optimal multi-point geostatistics. In the system 199, a user 102 can use a user computer or device to provide data that may include location, date, climate variables, and condition points, which is acquired using the system's 199 computer by an acquisition data function 200. The acquisition data function 200 can also collect data from a climate database 212 having climate variables, location, and date, which in one embodiment can be provided by the user. Climate variables can be precipitation measurements, temperature measurements, barometric pressure measurements, and other variables. Climate variables can include date, and location, latitude and longitude coordinates, which can be provided by the user using a computer or device. The user can also provide conditioning points, which can be retained from the original image for the simulation.

[0066] The system 199 receives the data to be simulated in an upscaling function 201 and processes the data reduction by upscaling, which generates smaller data applying direct sampling or parallel direct sampling, or any other multi-point geostatistical method. Various techniques may be used to determine the upscaling. For example, bicubic resampling or average pooling can be used as upscaling techniques.

[0067] In operation 201A, the system determines whether a CPU (Central Processing Unit) is available. If a CPU is available, the system proceeds to the parallel multi-point geostatistical method, as in function 205. If a CPU is not available, the system proceeds to function 202.

[0068] The reduced data processed by upscaling is received in function 202, which processes the upscaled data reduction using a multipoint geostatistical method to generate a simulation of the same size as the received data, for example using direct sampling or any other multipoint geostatistical method.

[0069] In operation 202A, the system determines whether the neural network is trained. If the neural network is trained, the system proceeds to the pre-trained model in function 204. If the neural network is not trained, the system proceeds to function 203.

[0070] Function 203 receives simulations generated by, for example, direct sampling or parallel direct sampling, or any other multi-point geostatistical method, and processes the scale increase by a downscaling method to generate data at the same scale as originally received.

[0071] Function 204 receives simulations performed by direct sampling or parallel direct sampling, or any other multi-point geostatistical method, and processes the scale increase with a neural network using a pre-trained model to generate data at the same scale as originally received. A super-resolution generative adversarial network can be used as the neural network.

[0072] Function 205 receives the reduced data and processes it using parallel direct sampling or any other multi-point geostatistical method to generate a simulation of the same size as the received data. Bicubic interpolation can be used as a downscaling technique.

[0073] Referring to FIG. 3 , in another embodiment according to the present disclosure, system 250 illustrates an offline system for preparing a neural network. System 250 illustrates an offline system and method for generating a pre-trained model for use in system 199. Function 206 collects relevant data for a given climate variable to prepare training and test data. Operation 206A determines whether the neural network is trained. The data includes a single variable having a certain size. Function 207 receives the reduced data and the original data to prepare training and test data. A combination of the reduced data and the augmented data is used to construct the training and test data sets. Function 208 receives the training data and performs fine-tuning of the neural network model. Function 209 receives the test data and evaluates the neural network model. System 250 determines when the neural network has achieved a sufficient accuracy rate in operation 209A. Various techniques may be used to determine the accuracy rate as root mean square error and absolute error. When the accuracy rate is sufficient, the system proceeds to function 204 with a completed pre-trained model. Other Examples and Embodiments

[0074] In another embodiment according to the present disclosure, a comparison can be made between the method of the present disclosure and the direct sampling method. The present invention can use 256x256 pixel precipitation data as training images, and in one example according to the present disclosure, simulations can be generated for this data. In one example, the samples represent a specific location on a certain day of the year, the climate variable is precipitation, and the method uses 3% of the images as conditioning points (points that the method should maintain).

[0075] The simulation can be generated using direct sampling, which, for an input image size of 256x256 pixels, takes 35,554 seconds, or roughly 10 hours of computation time, to generate the simulation.

[0076] Using a system including a super-resolution network according to the present disclosure, a simulation can be generated when the system has training data for the neural network to create the super-resolution samples. An upscaling technique using average pooling can be used, and a downscaling technique using a neural network SRGAN (a combination of a generative adversarial network (GAN) and a deep convolutional neural network (CNN)) trained on a dataset using an offline system can be used. In this scenario, with an input image size of 256x256 pixels, the computation time is 145 seconds, roughly 2 minutes, which is roughly a 245x speedup compared to the above example using direct sampling.

[0077] If the system does not have training data for that location, it can generate a simulation using the above techniques. A method based on bicubic interpolation techniques can be used for upscaling and downscaling. In this case, the calculation time is 141 seconds, or roughly 2 minutes, a 252-fold speed improvement compared to direct sampling methods. However, methods that do not use neural networks are of lower quality, as high frequencies are ignored in this case. In all examples, parallel methods can improve calculation time, which in this case can be reduced by a factor of 980. Additional Examples and Embodiments

[0078] 4 and 5 , in one embodiment according to the present disclosure, a method 600 for improving performance of a computer simulation using sampling can use the system 500. The method 600 includes processing, as in operation 604, using a computer 510 to reduce resolution 512 of input data 520 in a computer using an upscaling technique 532 that provides upscaling of the input data. In one example, the input data received at the computer can include variables. In another example, the input data can include weather-related information, including weather variables and weather conditions. In another example, the input data can include size and resolution, location, date, conditioning point, and weather variable data. In another example, the weather variables and conditions can be received from a computer 503 of a user 502 related to a location.

[0079] As in operation 604, upscaling the input data generates reduced resolution data 514 having a lower resolution and a smaller data size than the input data.

[0080] The input data has an input data size, and upscaling the input data produces a resolution reduction of the input data having a reduced resolution and a reduced data size, where the reduced data size is smaller than the input data size.

[0081] The method 600 includes applying a multi-point geostatistical technique 530 to the reduced-resolution data to generate a simulation 516 of a new weather scenario using a computer, as in operation 608, without modifying the reduced-resolution data 520 by reduced data size or reduced resolution. In one example, the simulation may be a digital weather simulation based on the reduced-resolution data.

[0082] The method 600 includes computationally processing the increased resolution of the simulated reduced resolution data using a downscaling technique 534 to return the reduced resolution data to the size and resolution of the input data, as in operation 612.

[0083] In one example, the simulation is a digital weather simulation based on reduced resolution data.

[0084] In another example, the input data includes location information including latitude and longitude, conditioning points, and weather-related information including weather information including weather conditions and dates.

[0085] In another example, weather variables and weather conditions are received from computer 503 of user 502 and associated with a particular location.

[0086] In one example, the generated simulation of the new weather scenario is the same data size as the reduced data size.

[0087] In another example, the multi-point geostatistical technique is a direct sampling method or a parallel direct sampling method. Alternatively, the sampling technique can be any other sampling technique that generates data with variability using the multi-point geostatistical technique 530.

[0088] In another example, the upscaling can be performed using a method that represents a point based on a set of points. For example, the method for upscaling can include bicubic resampling, average pooling, or another method that can represent a point based on a set of points.

[0089] In another example, downscaling (using downscaling technique 534) can be performed using a method to represent a set of points based on a neighborhood or based on a previous set of data associated with the same location or locations as the input data.

[0090] In another example, the method may further include using a neural network (as an example of a downscaling technique 534) to perform the downscaling if there is an available database and pre-trained neural network for the requested data, and if no neural network is available, upscaling may be performed using upscaling techniques 532 such as bicubic resampling or average pooling, and downscaling may be performed using an interpolation-based method using a single data sample.

[0091] In another example, the method may further include performing upsampling and downsampling using one or more methods other than a generative neural network if one or more neural networks are unavailable. If one or more neural networks are available with training data, performing upsampling and downsampling may use a generative neural network.

[0092] In another embodiment according to the present invention, a system for improving the performance of computer simulations using sampling includes a computer system comprising a computer processor, a computer-readable storage medium, and program instructions, the program instructions being stored on the computer-readable storage medium and executable by the processor to cause the computer system to perform the following functions: for processing resolution reduction of input data using an upscaling technique to provide upscaling of the input data, the input data including size and resolution, location, date, conditioning point, and meteorological variable data, and for upscaling the input data to generate reduced-resolution data having a reduced resolution and reduced data size that is lower resolution and smaller data size than the input data; for applying multi-point geostatistical techniques to the reduced data to generate simulations of new meteorological scenarios using a computer without making any reduced data size or reduced resolution changes to the reduced-resolution data; and for processing resolution increase of reduced-resolution data from a simulation using a downscaling technique to return the reduced-resolution data to the size and resolution of the input data.

[0093] In another embodiment according to the present disclosure, a computer program product for improving the performance of a computer simulation uses sampling. The computer program product comprises a computer-readable storage medium having program instructions embodied therein, the program instructions being executable by a computer to cause the computer to perform computer-implemented functions including the following functions: Function for using a computer to process resolution reduction of input data using an upscaling technique to provide upscaling of the input data; The input data includes size and resolution, location, date, and conditioning point, and weather variable data, and the upscaling of the input data generates reduced-resolution data having a reduced resolution and reduced data size that is lower resolution and smaller data size than the input data; Function for using a computer to apply multi-point geostatistical techniques to the reduced-resolution data to generate simulations of new weather scenarios without applying reduced data size or reduced resolution changes to the reduced-resolution data; Function for using a computer to process resolution increase of reduced-resolution data from a simulation using a downscaling technique to return the reduced-resolution data to the size and resolution of the input data. Additional Embodiments and Examples

[0094] Referring again to FIG. 4 , system 500 includes computer 572, which may be integral with or in communication with the device and may communicate with other computers, such as computer 510. A computer 590 remote from computer 510 may be in full or partial electronic communication with control system computer 572 as part of control system 570. Control system 570 may include computer 572 having a computer-readable storage medium 573 capable of storing one or more programs 574 and a processor 575 for executing program instructions. Control system 570 may include control software 578 for managing the one or more programs. The control system may also include a storage medium that may include registration and / or account data 582 and a user profile 583 for a user or entity (such an entity may include a robot entity) as part of a user account 581. User account 581 may be stored in storage medium 580 that is part of control system 570. User account 581 may include registration and account data 582 and a user profile 583. The control system may also include a computer 572 having a computer-readable storage medium 573 capable of storing a program or code embodied in the storage medium. The program code may be executed by a processor 575. The computer 572 may be in communication with a database 576. The control system 570 may also include a database 576 for storing all or some of the data as described above, as well as other data.

[0095] Control system 570 may also be in communication with computer 510, which may include learning engine / module 592 and knowledge corpus or database 596. Computer system 590 may also be in communication with computer 510 and may be remote from user 502 and computer 510. The depiction of computer 510 and other components of system 500 is shown as an example according to the present disclosure. One or more computer systems may be in communication with a communications network 560, e.g., the Internet. Further Embodiments and Examples

[0096] For example, account data, including profile data and personal or other data about a user, may be collected and stored, for example, in a control system. It is understood that such data collection is done with the user's knowledge and consent and stored in a manner that protects privacy, as discussed in more detail below. Such data may include personal data and data about personal items. With respect to the collection of data related to the present disclosure, such uploading or creation of a profile is optional by one or more users and is therefore initiated by a user with the user's approval. A user may thereby opt in to establishing an account with a profile according to the present disclosure. A user may thereby opt in to input data according to the present disclosure. Such user approval also includes the user's option to cancel such profile or account and / or data input, thereby opting out of communication and data capture at the user's discretion. It is further understood that any stored or collected data is intended to be securely stored, unavailable without the user's approval, and not available to the public or unauthorized users, or both. It is understood that such stored data will be deleted upon user request, and in a secure manner, and that any use of such stored data will be only with the user's approval and consent, in accordance with this disclosure. Further Examples and Embodiments

[0097] Additionally, methods and systems according to embodiments of the present disclosure may be discussed in terms of functional systems depicted by functional block diagrams. The methods and systems may include components and operations that are used herein to refer to when describing operational steps of the disclosed methods and systems for embodiments according to the present disclosure. Furthermore, a functional system according to an embodiment of the present disclosure depicts functional operations that are indicative of the embodiments discussed herein.

[0098] The methods and systems of the present disclosure may include a series of operational blocks for implementing one or more embodiments according to the present disclosure. The method shown in a figure may be another exemplary embodiment, which may include aspects / operations shown in another figure and discussed previously but which may be reintroduced in another example. As such, the operational blocks and system components shown in one or more of the figures may be similar to the operational blocks and system components in other figures. The variety of operational blocks and system components depicts exemplary embodiments and aspects according to the present disclosure. For example, the method shown is intended as an exemplary embodiment that may include aspects / operations shown and discussed previously in this disclosure, and in one example, continues from a previous method shown in another flowchart.

[0099] Features shown in some of the drawings, such as block diagrams, are understood to be functional representations of features of the present disclosure, which are shown in embodiments of the systems and methods of the present disclosure for illustrative purposes to clarify the functionality of the features of the present disclosure. Further Considerations Regarding Examples and Embodiments

[0100] A set or group is understood to be a collection of individual objects or elements. The objects or elements that make up a set or group can be anything, such as numbers, letters of the alphabet, other sets, multiple people or users, etc. It is further understood that a set or group can be a single element, such as a thing or number, or in other words, a set of elements, such as one or more users or people or participants. It is also understood that machine and device are used interchangeably herein to refer to machines or devices in a single, ecosystem, or environment, which may include, for example, an artificial intelligence (AI) environment.

[0101] The descriptions of various embodiments of the present invention have been presented for illustrative purposes and are not intended to be exhaustive or limited to the disclosed embodiments. Similarly, examples of features or functions of embodiments of the present disclosure described herein, whether used to describe a particular embodiment or listed as examples, are not intended to limit the embodiments of the present disclosure described herein or to limit the disclosure to the examples described herein. Such examples are intended to be examples or illustrative and not exhaustive. Many modifications and variations will be apparent to those skilled in the art without departing from the spirit and scope of the described embodiments. The terminology used herein has been selected to best explain the principles, practical applications, or technical improvements to technology found in the marketplace of the embodiments, or to enable others skilled in the art to understand the embodiments disclosed herein.

[0102] It is also understood that one or more computers or computer systems shown in the figures can include all or a portion of a computing environment and its components shown in another figure, e.g., computing environment 1000 can be incorporated in whole or in part into one or more computers or devices shown in other figures and described herein. In one example, one or more computers can communicate with all or a portion of the computing environment and its components as remote computer systems to achieve the computer functionality described in this disclosure. Further Examples and Embodiments

[0103] Various aspects of the present disclosure are described through text, flowcharts, block diagrams of computer systems, and / or block diagrams of machine logic included in embodiments of a computer program product (CPP). With respect to any flowchart, depending on the technology involved, operations may be performed in an order different from that shown in a given flowchart. For example, again depending on the technology involved, two operations shown in successive flowchart blocks may be performed in the reverse order, as a single integrated step, simultaneously, or in an at least partially overlapping manner.

[0104] A computer program product embodiment ("CPP embodiment" or "CPP") is a term used in this disclosure to describe any set of one or more storage media (also referred to as "media") collectively included in a set of one or more storage devices that collectively contain machine-readable code corresponding to instructions and / or data for performing the computer operations specified in a given CPP claim. A "storage device" is any tangible device that can hold and store instructions for use by a computer processor. The computer-readable storage medium may be, but is not limited to, an electronic storage medium, a magnetic storage medium, an optical storage medium, an electromagnetic storage medium, a semiconductor storage medium, a mechanical storage medium, or any suitable combination of the foregoing. Some known types of storage devices that include these media include diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital versatile disk (DVD), memory stick, floppy disk, mechanically encoded devices (such as punch cards or pits / lands formed on a major surface of a disk), or any suitable combination of the foregoing. Computer-readable storage media, as the term is used in this disclosure, is not to be construed as storage of transitory signals themselves in the form of radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide, light pulses passing through fiber optic cables, electrical signals communicated through wires, and / or other transmission media.As will be appreciated by those skilled in the art, data is typically moved at some infrequent time during the normal operation of a storage device, such as during access, defragmentation, or garbage collection, but the above does not make the storage device temporary, as the data is not temporary while it is stored.

[0105] 6, computing environment 1000 includes an example of an environment for executing at least some of the computer code involved in performing the methods of the present invention, such as improving performance of a computer simulation using sampling 1200. In addition to block 1200, computing environment 1000 includes, for example, computer 1101, wide area network (WAN) 1102, end user device (EUD) 1103, remote server 1104, public cloud 1105, and private cloud 1106. In this embodiment, computer 1101 includes a set of processors 1110 (including processing circuitry 1120 and cache 1121), a communications fabric 1111, volatile memory 1112, persistent storage 1113 (including operating system 1122 and block 1200, as identified above), a set of peripheral devices 1114 (including a set of user interface (UI) devices 1123, storage 1124, and a set of Internet of Things (IoT) sensors 1125), and a network module 1115. Remote server 1104 includes a remote database 1130. Public cloud 1105 includes a gateway 1140, a cloud orchestration module 1141, a set of host physical machines 1142, a set of virtual machines 1143, and a set of containers 1144.

[0106] Computer 1101 may take the form of a desktop computer, a laptop computer, a tablet computer, a smartphone, a smartwatch or other wearable computer, a mainframe computer, a quantum computer, or any other form of computer or mobile device now known or later developed that is capable of executing programs, accessing a network, or querying a database, such as remote database 1130. As is well understood in the art of computer technology, and depending on the technology, execution of a computer-implemented method may be distributed among multiple computers and / or among multiple locations. While in this presentation of computing environment 1100, to keep the presentation as simple as possible, the detailed discussion focuses on a single computer, specifically computer 1101. Computer 1101 may be located in a cloud, even though it is not depicted in the cloud of FIG. 6 . However, computer 1101 is not required to be in a cloud unless expressly indicated.

[0107] The processor set 1110 includes one or more computer processors of any type now known or later developed. The processing circuitry 1120 may be distributed across multiple packages, e.g., multiple tailored integrated circuit chips. The processing circuitry 1120 may implement multiple processor threads and / or multiple processor cores. The cache 1121 is memory located within the processor chip package and is typically used for data or code that should be available for fast access by threads or cores executing on the processor set 1110. Cache memory is typically organized into multiple levels depending on relative proximity to the processing circuitry. Alternatively, some or all of the cache for a processor set may be located “off-chip.” In some computing environments, the processor set 1110 may be designed to operate with qubits and perform quantum computing.

[0108] Computer-readable program instructions are typically loaded onto computer 1101 and cause processor set 1110 of computer 1101 to perform a series of operational steps, thereby realizing a computer-implemented method. As a result, the instructions so executed instantiate the method set forth in the flowcharts and / or descriptions of the computer-implemented method (collectively, the "invention methods") contained herein. These computer-readable program instructions are stored on various types of computer-readable storage media, such as cache 1121 and other storage media discussed below. The program instructions and associated data are accessed by processor set 1110 to control and direct the execution of the inventive methods. In computing environment 1100, at least some of the instructions for performing the inventive methods may be stored in persistent storage 1113 in block 1200.

[0109] Communications fabric 1111 is the signal-conducting pathway that allows the various components of computer 1101 to communicate with one another. Typically, this fabric is made up of switches and conductive pathways, such as those that make up buses, bridges, physical input / output ports, and the like. Other types of signal communication pathways may be used, such as fiber optic and / or wireless communication pathways.

[0110] Volatile memory 1112 may be any type of volatile memory now known or later developed. Examples include dynamic type random access memory (RAM) or static type RAM. Typically, volatile memory is characterized by random access, although this is not required unless explicitly stated. In computer 1101, volatile memory 1112 is located in a single package and internal to computer 1101, although alternatively or additionally, volatile memory may be distributed across multiple packages and / or located external to computer 1101.

[0111] Persistent storage 1113 is any form of non-volatile storage for a computer, now known or later developed. The non-volatility of this storage means that stored data remains regardless of whether power is supplied to computer 1101 and / or to persistent storage 1113 directly. Persistent storage 1113 may be read-only memory (ROM), but typically is at least a portion of persistent storage that allows data to be written, data to be erased, and data to be rewritten. Some well-known forms of persistent storage include magnetic disks and solid-state storage devices. Operating system 1122 may take several forms, such as various known proprietary operating systems or open-source Portable Operating System Interface-type operating systems, which utilize a kernel. The code contained in block 1200 typically includes at least a portion of the computer code involved in performing the methods of the present invention.

[0112] Peripheral device set 1114 includes a set of peripheral devices of computer 1101. Data communication connections between peripheral devices and other components of computer 1101 may be implemented in various manners, such as Bluetooth® connections, near field communication (NFC) connections, connections via cables (such as Universal Serial Bus (USB)-type cables), pluggable connections (e.g., Secure Digital (SD) cards), connections made over local area communication networks, and even connections made over wide area networks such as the Internet. In various embodiments, UI device set 1123 may include components such as display screens, speakers, microphones, wearable devices (such as goggles and smartwatches), keyboards, mice, printers, touchpads, game controllers, and haptic devices. Storage 1124 may be external storage such as an external hard drive or insertable storage such as an SD card. Storage 1124 may be persistent and / or volatile. In some embodiments, storage 1124 may take the form of a quantum computing storage device for storing data in the form of qubits. In embodiments where computer 1101 is required to have a large amount of storage (e.g., computer 1101 stores and manages a large database locally), this storage may be provided by a peripheral storage device designed to store very large amounts of data, such as a storage area network (SAN) shared by multiple geographically distributed computers. IoT sensor set 1125 consists of sensors that can be used in Internet of Things applications. For example, one sensor may be a thermometer and another sensor may be a motion detector.

[0113] The network module 1115 is a collection of computer software, hardware, and firmware that enables the computer 1101 to communicate with other computers over the WAN 1102. The network module 1115 may include hardware such as a modem or Wi-Fi® signal transceiver, software for packetizing and / or de-packetizing data for communication network transmission, and / or web browser software for communicating data over the Internet. In some embodiments, the network control and network forwarding functions of the network module 1115 are performed on the same physical hardware device. In other embodiments (e.g., embodiments utilizing Software-Defined Networking (SDN)), the control and forwarding functions of the network module 1115 are performed on physically separate devices, such that the control function manages multiple different network hardware devices. Computer-readable program instructions for carrying out the methods of the present invention may be downloaded to the computer 1101, typically from an external computer or external storage device, through a network adapter card or network interface included in the network module 1115.

[0114] WAN 1102 is any wide area network (e.g., the Internet) capable of communicating computer data over non-local distances using any technology for communicating computer data now known or later developed. In some embodiments, a WAN may be replaced and / or supplemented by a local area network (LAN) designed to exchange data between devices located in a local area, such as a Wi-Fi network. WANs and / or LANs typically include copper transmission cables, optical fiber transmissions, wireless transmissions, and computer hardware such as routers, firewalls, switches, gateway computers, and edge servers.

[0115] The end-user device (EUD) 1103 is any computer system used and controlled by an end user (e.g., a customer of the enterprise operating the computer 1101) and may take any of the forms discussed above in connection with the computer 1101. The EUD 1103 typically receives useful and useful data from the operation of the computer 1101. For example, in a hypothetical case in which the computer 1101 is designed to provide recommendations to the end user, the recommendations would typically be communicated from the network module 1115 of the computer 1101 over the WAN 1102 to the EUD 1103. In this manner, the EUD 1103 can display or otherwise present the recommendations to the end user. In some embodiments, the EUD 1103 may be a client device, such as a thin client, a heavy client, a mainframe computer, a desktop computer, or the like.

[0116] Remote server 1104 is any computer system that provides at least some data and / or functionality to computer 1101. Remote server 1104 may be controlled and used by the same entity that operates computer 1101. Remote server 1104 represents a machine that collects and stores useful and useful data for use by other computers, such as computer 1101. For example, in the hypothetical case where computer 1101 is designed and programmed to provide recommendations based on historical data, then this historical data may be provided to computer 1101 from remote database 1130 of remote server 1104.

[0117] A public cloud 1105 is any computer system available for use by multiple entities that provides on-demand availability of computer system resources and / or other computer capacity, particularly data storage (cloud storage) and computing capacity, without requiring direct, active management by users. Cloud computing typically leverages resource sharing to achieve coherence and economies of scale. Direct, active management of the computing resources of the public cloud 1105 is performed by computer hardware and / or software in a cloud orchestration module 1141. The computing resources provided by the public cloud 1105 are typically implemented by virtual computing environments running on various computers comprising a host physical machine set 1142, which is the universe of physical computers within and / or available in the public cloud 1105. A virtual computing environment (VCE) typically takes the form of virtual machines from a virtual machine set 1143 and / or containers from a container set 1144. It is understood that these VCEs may be stored as images and transferred among various physical machine hosts either as images or after instantiation of the VCE. Cloud orchestration module 1141 manages the transfer and storage of images, deploys new instantiations of VCEs, and manages active instantiations of VCE deployments. Gateway 1140 is a collection of computer software, hardware, and firmware that enables public cloud 1105 to communicate over WAN 1102.

[0118] Here, we provide some further explanation of virtual computing environments (VCEs). A VCE can be stored as an "image." From this image, a new, active instance of the VCE can be instantiated. Two well-known types of VCEs are virtual machines and containers. A container is a VCE that uses operating system-level virtualization. This refers to a feature of an operating system where the kernel allows the existence of multiple isolated user space instances called containers. These isolated user space instances typically behave as actual computers from the perspective of the programs running within them. A computer program running on a typical operating system can use all of the computer's resources, such as connected devices, files and folders, network shares, CPU power, and quantifiable hardware capabilities. However, a program running inside a container can only use the contents of the container and the devices assigned to the container, a feature known as containerization.

[0119] Private cloud 1106 is similar to public cloud 1105, except that its computing resources are available only for use by a single enterprise. While private cloud 1106 is shown in communication with WAN 1102, in other embodiments, the private cloud may be entirely disconnected from the Internet and accessible only through a local / private network. A hybrid cloud is a composite of multiple clouds of different types (e.g., private, community, or public cloud types), often each implemented by a different vendor. While each of the multiple clouds remains a separate, discrete entity, the larger hybrid cloud architecture is bound together by standardized or proprietary technologies that enable orchestration, management, and / or data / application portability between the constituent clouds. In this embodiment, both public cloud 1105 and private cloud 1106 are part of a larger hybrid cloud.

Claims

1. 1. A method for improving performance of a computer simulation using sampling, comprising: processing, using a computer, a resolution reduction of the input data by an upscaling technique that provides upscaling of the input data, the input data including data size and resolution, location, date, conditioning point, and weather variable data, wherein the upscaling of the input data generates reduced-resolution data having a reduced resolution and reduced data size that is lower in resolution and smaller in data size than the input data; applying multi-point geostatistical techniques to the reduced-resolution data to generate simulations of new weather scenarios using the computer without modifying the reduced data size or resolution of the reduced-resolution data; and processing, using the computer, the increased resolution of the reduced resolution data from the simulation using a downscaling technique to restore the reduced resolution data to the data size and resolution of the input data; A method for providing

2. The method of claim 1 , wherein the simulation is a digital weather simulation based on the reduced-resolution data.

3. The method of claim 1 , wherein the input data includes weather-related information including location information including latitude and longitude, conditioning points, and weather information including weather conditions and dates.

4. The method of claim 3 , wherein the weather variable data and the conditioning points are received from a user's computer and are associated with a particular location.

5. The method of claim 1 , wherein the generated simulation is the same size as the data size of the input data as received at the computer.

6. The method of claim 1 , wherein the multipoint geostatistical technique is a direct sampling method or a parallel direct sampling method.

7. The method of claim 1 , wherein the upscaling is performed using a method of representing one point based on a set of points.

8. The method of claim 1 , wherein the downscaling is performed using a method for representing a set of points based on a neighborhood or based on a previous set of data associated with the location.

9. If there is a database available and a pre-trained neural network is available for the requested data, using the neural network to perform the upscaling or downscaling; and If the neural network is not available, the upscaling is performed using bicubic resampling or average pooling, and the downscaling is performed using an interpolation-based method using one data sample. The method of claim 1 , further comprising:

10. If the neural network is unavailable, performing upsampling and downsampling using one or more methods other than a generative neural network; and performing upsampling and downsampling using the generative neural network if the neural network is available with training data; The method of claim 9 further comprising:

11. 1. A system for improving performance of a computer simulation using sampling, comprising: a computer system having a computer processor, a computer-readable storage medium, and program instructions stored on the computer-readable storage medium and installed on the computer system; a step of processing, using a computer, a resolution reduction of input data by an upscaling technique that provides upscaling of the input data, the input data including size and resolution, location, date, conditioning point, and weather variable data, the upscaling of the input data generating reduced-resolution data having a lower resolution and a smaller data size than the input data; applying multi-point geostatistical techniques to the reduced-resolution data to generate simulations of new weather scenarios using the computer without modifying the reduced-resolution data with respect to the reduced data size or the reduced resolution; and processing, using the computer, the increased resolution of the reduced-resolution data from the simulation using a downscaling technique to restore the reduced-resolution data to the size and resolution of the input data. A system executable by the computer processor to cause the computer to perform functions to perform the above.

12. The system of claim 11 , wherein the simulation is a digital weather simulation based on the reduced-resolution data.

13. The system of claim 11 , wherein the input data includes weather-related information including location information including latitude and longitude, conditioning points, and weather information including weather conditions and dates.

14. The system of claim 11 , wherein the weather variable data and the conditioning points are received from a user's computer and are associated with a particular location.

15. The system of claim 11 , wherein the generated simulation is the same size as the size of the input data as received at the computer.

16. The system of claim 11 , wherein the multipoint geostatistical technique is a direct sampling method or a parallel direct sampling method.

17. The system of claim 11 , wherein the upscaling is performed using a method of representing one point based on a set of points.

18. The system of claim 11 , wherein the downscaling is performed using a method for representing a set of points based on a neighborhood or based on a previous set of data associated with the location.

19. if a database is available and a pre-trained neural network is available for the requested data, using said neural network to perform said upscaling or said downscaling; and If the neural network is not available, the upscaling is performed using bicubic resampling or average pooling, and the downscaling is performed using an interpolation-based method using one data sample.

19. The system of claim 11, further comprising:

20. 1. A computer program for improving the performance of a computer simulation using sampling, comprising: and program instructions comprising: a step of processing, using a computer, a resolution reduction of input data by an upscaling technique that provides upscaling of the input data, the input data including size and resolution, location, date, conditioning point, and weather variable data, the upscaling of the input data generating reduced-resolution data having a lower resolution and a smaller data size than the input data; applying multi-point geostatistical techniques to the reduced-resolution data to generate simulations of new weather scenarios using the computer without modifying the reduced-resolution data with respect to the reduced data size or the reduced resolution; and processing, using the computer, the increased resolution of the reduced-resolution data from the simulation using a downscaling technique to restore the reduced-resolution data to the size and resolution of the input data. A computer program executable by a computer processor to cause the computer processor to execute the functions by a computer, including functions for performing the above.

Images