Schlumberger Technology Corporation (20240303925). GENERATING GEOLOGICAL FACIES MODELS WITH FIDELITY TO THE DIVERSITY AND STATISTICS OF TRAINING IMAGES USING IMPROVED GENERATIVE ADVERSARIAL NETWORKS simplified abstract

From WikiPatents
Jump to navigation Jump to search

GENERATING GEOLOGICAL FACIES MODELS WITH FIDELITY TO THE DIVERSITY AND STATISTICS OF TRAINING IMAGES USING IMPROVED GENERATIVE ADVERSARIAL NETWORKS

Organization Name

Schlumberger Technology Corporation

Inventor(s)

Lingchen Zhu of Cambridge MA (US)

Tuanfeng Zhang of Lexington MA (US)

GENERATING GEOLOGICAL FACIES MODELS WITH FIDELITY TO THE DIVERSITY AND STATISTICS OF TRAINING IMAGES USING IMPROVED GENERATIVE ADVERSARIAL NETWORKS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240303925 titled 'GENERATING GEOLOGICAL FACIES MODELS WITH FIDELITY TO THE DIVERSITY AND STATISTICS OF TRAINING IMAGES USING IMPROVED GENERATIVE ADVERSARIAL NETWORKS

Simplified Explanation

This patent application describes neural network systems and machine learning methods for geological modeling using an improved generative adversarial network. The network includes a generator neural network and a discriminator neural network to create simulated images of geological facies.

  • The generator neural network maps a noise vector and a category code vector to generate simulated images of geological facies.
  • The discriminator neural network determines the probability that an input image of geological facies is either a training image or a simulated image produced by the generator neural network.

Key Features and Innovation

  • Utilizes a generative adversarial network for geological modeling.
  • Combines a generator neural network and a discriminator neural network.
  • Maps noise and category code vectors to create simulated images of geological facies.
  • Determines the probability of input images being training or simulated images.

Potential Applications

This technology can be applied in various fields such as geology, mining, and environmental science for creating realistic geological models.

Problems Solved

  • Improves the accuracy and efficiency of geological modeling.
  • Enables the generation of realistic images of geological facies.
  • Enhances the training process for neural networks in geological applications.

Benefits

  • Streamlines the process of creating geological models.
  • Increases the quality of simulated images.
  • Enhances the understanding of geological formations.

Commercial Applications

  • Geological surveying and exploration companies can use this technology to improve their modeling capabilities.
  • Environmental agencies can benefit from more accurate geological assessments for land management.

Prior Art

Further research can be conducted on existing generative adversarial networks and their applications in geological modeling to identify any related prior art.

Frequently Updated Research

Researchers are continually exploring new techniques and advancements in neural networks for geological modeling, which may impact the development of this technology.

Questions about Geological Modeling

How can this technology improve the accuracy of geological predictions?

This technology enhances the generation of realistic geological facies images, leading to more precise predictions based on the data.

What are the potential limitations of using generative adversarial networks in geological modeling?

Generative adversarial networks may require significant computational resources and data to train effectively, which could be a limitation in certain geological modeling applications.


Original Abstract Submitted

neural network systems and related machine learning methods for geological modeling are provided that employ an improved generative adversarial network including a generator neural network and a discriminator neural network. the generator neural network is trained to map a combination of a noise vector and a category code vector as input to a simulated image of geological facies. the discriminator neural network is trained to map at least one image of geological facies provided as input to corresponding probability that the at least one image of geological facies provided as input is a training image of geological facies or a simulated image of geological facies produced by the generator neural network.

Retrieved from "http://wikipatents.org/index.php?title=Schlumberger_Technology_Corporation_(20240303925)._GENERATING_GEOLOGICAL_FACIES_MODELS_WITH_FIDELITY_TO_THE_DIVERSITY_AND_STATISTICS_OF_TRAINING_IMAGES_USING_IMPROVED_GENERATIVE_ADVERSARIAL_NETWORKS_simplified_abstract&oldid=227174"