Digital Twin Network Mapping for Power Prediction Gaps

Solution Overview

Problem

Artificial intelligence models face challenges in adapting to environments with limited or no data, particularly in complex network arrangements with non-homogenous or unknown device specifications, making it difficult to predict power consumption and carbon emissions effectively.

Innovation Solution

The system generates a digital twin of devices with non-homogenous or unknown specifications by creating a network mapping and inferring virtual specifications, allowing for the completion of incomplete data sets and proper training of AI models to provide power consumption predictions and recommend optimal network arrangements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If artificial intelligence models are used to predict power consumption in network arrangements, then prediction capability is improved, but the requirement for large amounts of high-quality training data creates implementation barriers in environments with limited or no available data

Engineering Contradiction:

Improvepower consumption prediction accuracyVSAvoidtraining data availability

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent creates virtual device profiles that copy and simulate the characteristics of physical devices with unknown specifications. These virtual profiles serve as synthetic training data, allowing the AI model to learn power consumption patterns without requiring actual measurement data from every device type. The virtual profiles replicate device behaviors and power characteristics based on available information and relationships within the network.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system performs preliminary data preparation by creating virtual device profiles and completing incomplete data sets before training the AI model. This advance preparation of training data ensures that the model has sufficient high-quality information to learn effective power consumption predictions, even when actual device data is limited or unavailable.

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If complete device specifications are required for accurate power consumption prediction, then prediction accuracy is improved, but the complexity of handling non-homogenous devices with unknown specifications increases

Engineering Contradiction:

Improvepower consumption prediction accuracyVSAvoidnetwork arrangement complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces virtual device profiles as intermediary representations between the AI model and physical devices with unknown specifications. These virtual profiles act as mediators that translate incomplete or non-homogenous device information into a standardized format that the AI model can process effectively, without requiring direct complete specifications from every physical device.

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system changes the parameter representation by creating virtual specifications that infer missing device parameters based on available information and device relationships. Instead of requiring all original device parameters to be known, the system transforms the problem by generating derived parameters through virtual profiling, allowing the AI model to work with completed parameter sets.

Inventive Principle:

Principle #35Parameter changes

3Productivity

If AI models are trained on available data, then model training is enabled, but incomplete data sets result in suboptimal prediction performance

Engineering Contradiction:

Improvemodel training capabilityVSAvoidprediction accuracy

Core Design Contradiction:

ProductivityVSMeasurement precision

Solution Approach 1:

The patent creates virtual device profiles that copy and simulate the characteristics of physical devices with unknown specifications. These virtual profiles serve as synthetic training data, allowing the AI model to learn power consumption patterns without requiring actual measurement data from every device type. The virtual profiles replicate device behaviors and power characteristics based on available information and relationships within the network.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The system performs preliminary data preparation by creating virtual device profiles and completing incomplete data sets before training the AI model. This advance preparation of training data ensures that the model has sufficient high-quality information to learn effective power consumption predictions, even when actual device data is limited or unavailable.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11843248B1Systems and methods for providing power consumption predictions for selected applications within network arrangements featuring devices with non-homogenous or unknown specifications

Publication Date: 2023.12.12 CITIBANK N A

AI summary

Systems and methods are described herein for novel uses and/or improvements to artificial intelligence applications in an environment with limited or no available data. In particular, systems and methods are described herein for providing network arrangement recommendations based on power consumption predictions for selected applications within network arrangements featuring devices with non-homogenous or unknown specifications.