Magnetic Coil Layout Optimization Using QUBO Flux Maximization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies fail to efficiently optimize the change amount of interlinkage magnetic flux in magnetic devices due to limitations in shape and arrangement optimization, particularly when using continuous design parameters and the annealing method, which restricts the ability to express objective functions in suitable formats for efficient optimization.
Innovation Solution
The proposed solution involves an optimization processing unit that divides the magnetic device's surface into coil regions, using auxiliary variables to express the existence of clockwise and counterclockwise coils, allowing for the optimization of interlinkage magnetic flux without shape or position limitations, and expressing the objective function in a Quadratic Unconstrained Binary Optimization (QUBO) format for efficient computation using the annealing method.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If continuous design parameters are used to optimize the shape of the coil, then the optimization can cover a wide range of values, but the shape is limited by the properties of the design parameter (e.g., circular shape when radius is the parameter)
Solution Approach 1:
The patent segments the coil shape optimization into discrete unit cells arranged in a grid pattern. Each unit cell can be independently configured to exist or not exist, allowing the construction of complex coil shapes without being constrained by continuous parameter formulations. This segmentation enables versatile shape optimization while avoiding the shape limitations inherent in continuous parameter approaches.
2Productivity
If the objective function is expressed with a small number of continuous design parameters, then optimization can be performed, but it is not possible to express the objective function in combination optimization problem form, limiting efficiency when degree of freedom is increased
Solution Approach 1:
The patent transforms the optimization problem from continuous parameters to discrete binary parameters. Each unit cell configuration is represented by binary variables (e.g., xij = 1 if clockwise coil exists, xij = 0 otherwise), converting the objective function into a combinatorial optimization problem suitable for annealing methods. This parameter transformation enables efficient optimization with high degrees of freedom while maintaining problem tractability.
3Loss of time
If the annealing method is used to optimize the objective function, then efficient optimization can be performed in short time, but the objective function must be expressed in a specific format with discretized parameters
Solution Approach 1:
The patent divides the coil design space into discrete unit cells that can be independently optimized. This segmentation naturally aligns with the annealing method's requirement for discretized parameters, enabling efficient computation without losing design flexibility. Each unit cell's binary state (present/absent, clockwise/counterclockwise) provides the discrete structure needed for annealing while maintaining sufficient design freedom.
Solution Approach 2:
The patent changes the parameter representation from continuous to discrete binary variables, making the objective function compatible with annealing optimization. This parameter transformation enables the use of efficient annealing algorithms that can quickly find optimal solutions, reducing optimization time significantly compared to continuous parameter methods.
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
This approach enables sufficient and efficient optimization of interlinkage magnetic flux by maximizing the sum of change amounts in each coil region, allowing for the optimization of coil shape, position, and magnet arrangement, thereby enhancing the performance of magnetic devices like vibration power generation devices.
Implementation Method 1
Because these magnetic devices use a physical phenomenon called electromagnetic induction, in order to improve performance of the magnetic device, for example, there is a case where it is requested to increase a change amount of a magnetic flux interlinking with (passing through) the coil generated from the magnet.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An optimizer that optimizes a change amount of an interlinkage magnetic flux in a coil of a magnetic device, the optimizer includes: an optimization processing unit configured to optimize the change amount of the interlinkage magnetic flux in the coil by using an objective function formula that maximizes a sum of Δϕirightxi (i = 1 to Nc) and Δϕileftyi (i = 1 to Nc), when it is assumed that a surface of the magnetic device where the coil is arranged be divided into Nc (Nc is integer) coil regions, in an i-th coil region Ni, an auxiliary variable of a clockwise coil that may exist in the coil region Ni be xi, and an auxiliary variable of a counterclockwise coil that may exist in the coil region Ni be yi, a case where the clockwise coil exists be xi = 1, a case where the clockwise coil does not exist be xi = 0, a case where the counterclockwise coil exists be yi = 1, a case where the counterclockwise coil does not exist be yi = 0, both of the clockwise coil and the counterclockwise coil do not exist when xi = 1 and yi = 1 are satisfied, and a change amount of an interlinkage magnetic flux of the clockwise coil in the coil region Ni be Δϕiright and a change amount of an interlinkage magnetic flux of the counterclockwise coil in the coil region Ni be Δϕileft.