In this thesis we present the optimisation of computer-synthesised electronic circuits.
We describe the field of genetic programming and its inspiration, taken from nature, with
emphasis on the evolutionary process and genetic operations. We take special note of the
subfield of grammatical evolution. Then we describe the application of genetic algorithms for
circuit synthesis, with the focus on different circuit representations, including grammatical
evolution.
We analyse the problem of bloat in synthesised circuits and introduce the theoretical
background of optimisation for circuits consisting of resistors, capacitors and inductors. We
describe our program for circuit optimisation written in the programming language Python.
We conclude by testing the program and present some examples of optimised circuits. We also
evaluate the capabilities and limitations of our solution.
|
