AI Shape Design for Wireless Power Transmission Coils
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless power transmission systems face challenges in designing optimal core and coil shapes to maximize efficiency and power transmission, as current methods rely on intuitive design and simulation, making it difficult to achieve the highest coupling coefficient.
Innovation Solution
A shape design system and method using machine learning to optimize core and coil shapes by performing learning based on shape information and performance evaluation, adjusting component values to improve wireless power transmission performance metrics such as magnetic flux density and coupling coefficient.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If intuitive design and simulation methods are used to design ferrite core shape, then design process can be completed with simple tools, but it is difficult to achieve optimal coupling coefficient and wireless power transmission efficiency
Solution Approach 1:
The patent replaces the traditional mechanical/intuitive design process with an artificial intelligence-based system. The AI model learns optimal ferrite core shapes through training on simulation data, substituting human designer intuition with automated machine learning algorithms that can objectively optimize coupling coefficients without requiring complex manual iteration.
Solution Approach 2:
The patent performs preliminary actions by pre-training the AI model with extensive simulation data before actual design. The system pre-learns the relationship between ferrite core shapes and coupling coefficients through offline training, so that when actual design is needed, the optimized shape can be quickly determined without time-consuming trial and error.
2Manufacturing precision
If extensive simulation and trial-and-error are performed to optimize core shape, then optimal coupling coefficient can be achieved, but design time and computational resources are significantly increased
Solution Approach 1:
The patent performs preliminary actions by pre-training the AI model with extensive simulation data before actual design. The system pre-learns the relationship between ferrite core shapes and coupling coefficients through offline training, so that when actual design is needed, the optimized shape can be quickly determined without time-consuming trial and error.
Solution Approach 2:
The patent creates a virtual copy of the design process through AI modeling. Instead of repeatedly performing physical simulations and trials, the system creates an AI model that copies and generalizes the design knowledge from training data, allowing rapid prediction of optimal shapes for new design scenarios without repeating the full simulation cycle.
3Productivity
If designer intuition is relied upon for ferrite core design, then design process is simple and fast, but it is difficult to ensure optimal shape for maximum power transmission
Solution Approach 1:
The patent replaces the traditional mechanical/intuitive design process with an artificial intelligence-based system. The AI model learns optimal ferrite core shapes through training on simulation data, substituting human designer intuition with automated machine learning algorithms that can objectively optimize coupling coefficients without requiring complex manual iteration.
Solution Approach 2:
The patent implements feedback mechanisms where the AI model is trained on simulation results that provide feedback on the relationship between ferrite core shapes and coupling coefficients. The system continuously learns from performance data, adjusting its predictions to improve accuracy, and can evaluate design quality through performance metrics before final manufacturing.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system effectively evaluates and optimizes core and coil shapes, enhancing wireless power transmission efficiency and power transfer by continuously learning and adjusting shape information until predetermined performance criteria are met, resulting in improved coupling coefficients and overall system performance.
Implementation Method 1
a ferrite core is used to increase a collection speed of the magnetic field and change a distribution of the magnetic field
Implementation Method 2
a ferrite core is used to increase a collection speed of the magnetic field
Implementation Method 3
a wireless power transmission device transmits power to a power receiving device by using an electromagnetic induction method
Data Source
AI summary
Proposed is a shape design technology for a wireless power transmission system, the shape design system including: a learning module configured to perform learning based on shape information and compensation information input in relation to a design target and generate new shape information; and an analysis module configured to evaluate wireless power transmission performance based on the shape information from the learning module and provide the learning module with a performance evaluation result.