Fujitsu Harnesses Quantum-Inspired Digital Annealer Technology to Optimize Output Efficiency in Power Generation for Renewable Energy Devices

An image in which the maximum plan of magnets optimizes the flux density towards the coil

Fujitsu Laboratories, Ltd. has actually revealed the ability of the Digital Annealer, Fujitsu’s computational architecture motivated by quantum phenomena that quickly resolves combinatorial optimization issues, to optimize the efficiency of magnetic devices important for renewable resource harvesting and other usages. The application of Fujitsu’s next-generation architecture enables for the almost instant estimation of the optimum plan of several planar (2D) magnets to optimize the strength of the electromagnetic field in a gadget.

Numerous magnetic devices utilized for ecological power generation develop magnetic flux through the plan of a a great deal of little magnets. The optimum planar (2D) plan for taking full advantage of power generation efficiency stays hard to compute due to the massive variety of prospective mixes of magnet plans, nevertheless. To conquer this difficulty, Fujitsu has actually established a technology that uses its Digital Annealer to compute in a matter of seconds how to set up each specific magnet to accomplish optimum magnetic flux density, providing an efficiency gain of 16%.

This technological development now makes it possible to rapidly compute the optimum style for magnetic devices with substantially greater power generation efficiency, and will eventually contribute to the spread of power generation devices that make use of renewable resource such as energy harvesting devices.

This technology was established in cooperation with Teacher Hajime Igarashi of Hokkaido University’s Institute of Details Science, and an abstract describing the technology will exist at the COMPUMAG 2019 (The 22nd International Conference on the Calculation of Electromagnetic Fields) conference held in Paris, France from Monday, July 15th.


Energy gathering devices, which transform vibrations from motors, engines, bridges and structures into electrical power, are significantly drawing in attention. The technology underpinning these devices removes the requirement for power transmission cable televisions, battery replacement and charging as a way of providing electrical power to IoT devices set up both inside your home and outdoors, in addition to for on-board power supply for wearable devices and vehicle elements. While embracing this technology provides an appealing service in a range of useful contexts, even more enhancements in the efficiency of energy harvesting devices will show important to fixing the higher ecological and energy issues dealing with mankind.


Energy gathering devices transform vibrations into electrical power utilizing the physical phenomenon of electro-magnetic induction produced by irreversible magnets and coils. To optimize the power generation efficiency of an energy harvesting gadget, the magnitude of the magnetic flux density originating from the lots of magnets situated within the gadget need to be optimized relative to the place of the coil.

At present, the design in which a variety of magnets is set up in a row (One Measurement) with a concentration of magnetic flux on one side is popular. Organizing the magnets in a planar shape (2D) will show reliable in increasing the quantity of power created, leading to the accomplishment of devices with even higher efficiency in the future. Given that the plan of the magnets set up in a planar shape (2D) is made complex, nevertheless, designers deal with a powerful difficulty when browsing for the optimum plan of magnets that optimizes flux density near the coil. In reality, the variety of possible mixes of magnet orientations when 10 × 10 magnets are set up in a square shape along a 3- dimensional coordinate axis is more than 77 to the power of 10.

The Newly-Developed Technology

Fujitsu has actually started a cooperation with Teacher Hajime Igarashi of Hokkaido University’s Institute of Details Science to establish a method for computing the optimum plan of planar magnets utilizing the Digital Annealer, which is a unique computing architecture that imitates quantum habits to quickly resolve combinatorial optimization issues not possible for standard computer systems.

For the Digital Annealer, issues need to be specified over binary (0 and 1) variables. In this research study, the instructions of a magnet that can be oriented along the 3 axes of X, Y, and Z is revealed by 3 bit variables, and the magnetic flux density created is developed utilizing the variable and Bio-Savart’s law, among the laws of electromagnetism, and an unbiased function (function whose worth need to be optimized) as a combinatorial optimization issue in which the magnetic flux density is optimized for a particular part. Moreover, by including a brand-new variable to the unbiased function so that it can be developed in the QUBO(1) format, the Digital Annealer it can compute the maximum style structure for planar magnet selections.


Utilizing Fujitsu’s Digital Annealer, it was possible to compute the maximum style of a planar (2D) magnet range from a large variety of prospective mixes. The simulation validated that the style optimization issue of 10 × 10 2-dimensional magnet selections can be resolved in a couple of seconds. By utilizing the resulting range, we were able to enhance the magnetic flux density by 17% and the power generation efficiency of the energy harvesting gadget by 16% compared to the conventionally-designed 2-dimensional range. This technology is likewise anticipated to be used to the optimization of magnet selections for direct motors, where magnetic flux density need to be managed as planned for greater efficiency.

Future Strategies

Fujitsu Laboratories will contribute to the additional advancement of magnetic devices utilized in energy harvesting, etc., by executing this technology as one of the expert services for Digital Annealer in financial 2020.

Recommended For You

About the Author: livescience

Leave a Reply

Your email address will not be published. Required fields are marked *