Transformation of Transformer Design with Artificial Intelligence

Abstract:
In the process of designing magnetic components, more than forty different parameters can impact the final design solution. Most of these design parameters can take over two hundred possible values each, altogether making countless number of design permutations for a particular set of input parameters. This makes it difficult to arrive at an optimum magnetic design. A human designer with expert knowledge and experience can eliminate many of these possible design permutations. However, the opinion that magnetic experts have on the impact of each of these parameters can be different. This means that the design output will highly depend on the knowledge and the experience of the designer. If it was possible to carry out an exhaustive evaluation of all the parameters with the use of a computer that would give a great advantage in arriving at the most optimized solution. However, the theoretical number of possible design permutations exceed the numbers of atoms in the world which makes the use of conventional computational and programming approaches impossible. A sophisticated Artificial Intelligence based autonomous approach was developed with the use of artificial intelligence algorithms to carry out the design process of magnetics and to arrive at designs that exceed even the superhuman designer. This process is discussed in this paper.