The semiconductor industry uses lithography machines for manufacturing complex integrated circuits (also called ICs) onto wafers. Because an IC is built up layer by layer it is very important to align the wafer each time it enters the machine. Today, gratings are used for this alignment step. Gratings are tiny periodic structures printed on a wafer and are even smaller than ICs. These gratings ...

We develop a forward modelling algorithm for computing time domain electromagnetic responses for conductive, permeable bodies in the low frequency regime. Both physical properties can be highly discontinuous. The parameter regimes considered give rise to highly stiff problems in the time domain and an implicit method, backward Euler, is used for time-stepping. At each time step the electromagne...

Servo control is usually done by means of model-based feedback controllers, which has two difficulties. Firstly, the design of a well performing feedback controller requires extensive and time consuming modelling of the process. Secondly, by applying feedback control a compromise has to be made between performance and robust stability. The learning feed forward controller (LFFC) may help to ove...

INTRODUCTION Groundwater plays an important part in many regolith processes (chemical weathering, solute dispersion, etc.), but collecting groundwater data usually requires drilling that can be: a) expensive analyses for porosity, permeability, etc in addition to drilling costs; b) problematic rig availability and access to remote sites; c) under-sampled entire groundwater regimes characterised...

The Kadomtsev-Petviashvili equation describes nonlinear dispersive waves which travel mainly in one direction, generalizing the Korteweg de Vries equation for purely uni-directional waves. In this paper we derive an improved KP-equation that has exact dispersion in the main propagation direction and that is accurate in second order of the wave height. Moreover, different from the KP-equation, t...

Parameter identification for wildfire forecasting models often relies on case-by-case tuning or posterior diagnosis/analysis, which can be computationally expensive due to the complexity of forward prediction model. In this paper, we introduce an efficient parameter flexible fire algorithm based machine learning and reduced order modelling techniques. Using a training dataset generated by physi...

We present a technique which uses forward modelling in conjunction with a genetic algorithm to nd the internal rotation proole which best reproduces a set of input frequency splittings. The genetic algorithm provides an eecient and robust method for global optimization. We evaluate the ability of this technique to determine the rotation proole in the solar core by applying it to artiicial datas...