Efficiency optimization of electrical devices

A Robust Bayesian Optimization Framework for Microwave Circuit Design under Uncertainty

“…In [39], an UQ-based optimization strategy, leveraging on knowledge-based feature surrogates, is introduced for the robust design of microwave components. For a broad overview of several robust optimization approaches, the reader is referred to [2,16,19,40].…”

“…In Section 2.2, it is described that the SE is used as a kernel of the GP model: the expectations in Equation ( 5) can be calculated analytically for the SE kernel. To obtain a sufficiently good approximation of the real distribution, in practice, we suggest to adopt only the mean and variance and calculate them through the law of total cumulance [55]: 2 . These stochastic moments are matched with a Gaussian distribution to approximate p( f |x , D n ).…”

“…The second microwave example is a zigzag narrow bandpass filter (see Figure 8) based on the design proposed in [58]. The filter is defined over the frequency range [2,3] GHz on a substrate with height h = 0.5 mm, relative permittivity r = 2.2 and loss tangent tanδ = 0.003. The width of both the horizontal and vertical conductors is 0.4 mm.…”

“…The emergence of new technologies and the constant miniaturization of integrated circuits challenge engineers to obtain designs that satisfy stringent functional specifications and signal integrity (SI) requirements, as well as electromagnetic compatibility (EMC) constraints. Various optimization techniques are used to find optimal designs and improve the device performance at the early design stage [1][2][3]. While in other engineering domains it is sometimes possible to solve complex numerical equations with efficient-stable numerical methods [4][5][6]; given the high computational cost of simulating modern high-frequency circuits, surrogate modeling techniques are typically utilized to efficiently perform design optimization [7][8][9][10][11].…”