Neural Flow-Field Sensing Without Added Physical Sensors

Solution Overview

Problem

Computing devices face limitations in obtaining fine-grained information due to the limited number and type of sensors, particularly lacking sensors for measuring parameters like pressure and flow rate, which hinders performance improvement and increases cost and size when additional sensors are added.

Innovation Solution

Utilizing a trained neural network model based on computational fluid dynamics (CFD) simulation data to determine flow field parameters such as temperature, pressure, and flow rate without additional physical sensors, allowing for flexible and cost-effective sensing of operating parameters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If additional physical sensors are added to measure parameters like pressure and flow rate, then measurement precision is improved, but device complexity and cost increase

Engineering Contradiction:

Improveflow field parameter measurementVSAvoidsensor quantity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent creates a virtual copy of the physical sensing system by training a neural network model on CFD simulation data. This digital twin replicates the functionality of physical pressure and flow rate sensors without requiring actual hardware sensors, thereby achieving measurement precision while avoiding increased device complexity

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical/physical sensor system with a computational intelligence system. Instead of using physical sensors to directly measure flow field parameters, the system uses a neural network model that processes temperature field data from existing sensors to infer pressure and flow rate parameters, substituting mechanical measurement with computational analysis

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If additional physical sensors are added to measure parameters like pressure and flow rate, then measurement precision is improved, but cost increases

Engineering Contradiction:

Improveflow field parameter measurementVSAvoiddevice cost

Core Design Contradiction:

Measurement precisionVSEase of manufacture

Solution Approach 1:

The patent creates a virtual copy of the physical sensing system by training a neural network model on CFD simulation data. This digital twin replicates the functionality of physical pressure and flow rate sensors without requiring actual hardware sensors, thereby achieving measurement precision while avoiding increased device cost

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces expensive physical sensors with a computationally-generated model that can be replicated at minimal cost. The neural network model, once trained on CFD simulation data, provides continuous flow field parameter estimation without the recurring costs of physical sensor manufacturing, calibration, and replacement

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

3Measurement precision

If additional physical sensors are added to measure parameters like pressure and flow rate, then measurement precision is improved, but device size increases

Engineering Contradiction:

Improveflow field parameter measurementVSAvoiddevice size

Core Design Contradiction:

Measurement precisionVSVolume of moving object

Solution Approach 1:

The patent creates a virtual copy of the physical sensing system by training a neural network model on CFD simulation data. This digital twin replicates the functionality of physical pressure and flow rate sensors without requiring actual hardware sensors, thereby achieving measurement precision while avoiding increased device size

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent transitions from physical spatial occupation to computational dimensional space. Instead of adding physical sensors that occupy device volume, the system adds a computational layer that processes data from existing sensors, moving the sensing function from the physical dimension to the digital dimension where no additional physical space is required

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS20250335763A1Method, electronic device, and computer program product for determining flow field parameter of object

Publication Date: 2025.10.30 DELL PROD LP

AI summary

A method in an illustrative embodiment includes: acquiring a heating parameter and a cooling parameter; and determining, based on the heating parameter and the cooling parameter, a flow field parameter at a target location in an object utilizing a trained neural network model, wherein the flow field parameter includes at least one of temperature, pressure, and flow rate at the target location in the object, and the trained neural network model is trained based on computational fluid dynamics (CFD) simulation sample data. By the method according to embodiments of the present disclosure, fluid parameters at the target location in the object can be determined using the trained neural network model, so that a variety of fine-grained information including temperature, flow rate, pressure, and the like can be obtained without additional physical sensors, and the cost of devices can also be saved.