Roughness sub-layer wind speed model for tropical wooded areas

There are few reports in the literature regarding wind speed near the ground. This work presents a model for wind speed from 4 m above the ground, based on year-round measurements in two meteorological towers. Each tower is equipped with anemometers at five heights, as well as thermometers and pressure and relative humidity sensors. The data is processed using Eureqa artificial intelligence software, which determines the functional relationship between variables using an evolutionary search technique called symbolic regression. Using this technique, models are found for each month under study, in which height and temperature are the variables that most affect wind speed. The model that best predicts the measured wind speeds is then selected. A polynomial function directly proportional to height and temperature is identified as the one that provides the best predictions of wind speed on average, within the rough sub-layer. Finally, future work is identified on testing the model at other locations.

The RBF Hyperparameter in Metamodel-Assisted Evolutionary Search

RBF metamodels, which are commonly used in expensive optimization problems, rely on a hyperparameter which affects their prediction. The optimal hyperparameter value is typically unknown and hence needs to be estimated by additional procedures. As such this study examines if this overhead is justified from an overall search effectiveness perspective, namely, if changes in the hyperparameter yield significant performance differences. Analysis based on extensive numerical experiments shows that changes are significant in functions with low to moderate multimodality but are less significant in functions with highly multimodality.

Constant optimization and feature standardization in multiobjective genetic programming

This paper extends the numerical tuning of tree constants in genetic programming (GP) to the multiobjective domain. Using ten real-world benchmark regression datasets and employing Bayesian comparison procedures, we first consider the effects of feature standardization (without constant tuning) and conclude that standardization generally produces lower test errors, but, contrary to other recently published work, we find much less clear trend for tree sizes. In addition, we consider the effects of constant tuning – with and without feature standardization – and observe that (1) constant tuning invariably improves test error, and (2) usually decreases tree size. Combined with standardization, constant tuning produces the best test error results; tree sizes, however, are increased. We also examine the effects of applying constant tuning only once at the end a conventional GP run which turns out to be surprisingly promising. Finally, we consider the merits of using numerical procedures to tune tree constants and observe that for around half the datasets evolutionary search alone is superior whereas for the remaining half, parameter tuning is superior. We identify a number of open research questions that arise from this work.

Automatic generation of atomic multiplicity-preserving search operators for search-based model engineering

Recently, there has been increased interest in combining model-driven engineering and search-based software engineering. Such approaches use meta-heuristic search guided by search operators (model mutators and sometimes breeders) implemented as model transformations. The design of these operators can substantially impact the effectiveness and efficiency of the meta-heuristic search. Currently, designing search operators is left to the person specifying the optimisation problem. However, developing consistent and efficient search-operator rules requires not only domain expertise but also in-depth knowledge about optimisation, which makes the use of model-based meta-heuristic search challenging and expensive. In this paper, we propose a generalised approach to automatically generate atomic multiplicity-preserving search operators for a given optimisation problem. This reduces the effort required to specify an optimisation problem and shields optimisation users from the complexity of implementing efficient meta-heuristic search mutation operators. We evaluate our approach with a set of case studies and show that the automatically generated rules are comparable to, and in some cases better than, manually created rules at guiding evolutionary search towards near-optimal solutions.