The intermittency of wind remains the greatest challenge to its large scale adoption and sustainability of wind farms. Accurate wind power predictions therefore play a critical role for grid efficiency where wind energy is integrated. In this paper, we investigate two hybrid approaches based on the genetic algorithm (GA) and particle swarm optimisation (PSO). We use these techniques to optimise an Adaptive Neuro-Fuzzy Inference system (ANFIS) in order to perform one-hour ahead wind power prediction. The results show that the proposed techniques display statistically significant out-performance relative to the traditional backpropagation least-squares method. Furthermore, the hybrid techniques also display statistically significant out-performance when compared to the standard genetic algorithm.

The greatest concern facing renewable energy sources like wind is the uncertainty in production volumes as their generation ability is inherently dependent on weather conditions. When providing forecasts for newly commissioned wind farms there is a limited amount of historical power production data, while the number of potential features from different weather forecast providers is vast. Bayesian regularization is therefore seen as a possible technique for reducing model overfitting problems that may arise. This work investigates Bayesian Neural Networks for one-hour ahead forecasting of wind power generation. Initial results show that Bayesian Neural Networks display equivalent predictive performance to Neural Networks trained by Maximum Likelihood. Further results show that Bayesian Neural Networks become superior after removing irrelevant features using Automatic Relevance Determination(ARD).

In recent years, wind and other variable renewable energy sources have gained a rapidly increasing share of the global energy mix. In this context the greatest concern facing renewable energy sources like wind is the uncertainty in production volumes as their generation ability is inherently dependent on weather conditions. When providing forecasts for newly commissioned wind farms there is a limited amount of historical power production data, while the number of potential features from different weather forecast providers is vast. Bayesian regularization is therefore seen as a possible technique for reducing model overfitting problems that may arise. This thesis investigates Bayesian Neural Networks in one-hour and day-ahead forecasting of wind power generation. Initial results show that Bayesian Neural Networks display equivalent predictive performance to Neural Networks trained by Maximum Likelihood in both one-hour and day ahead forecasting. Models selected using maximum evidence were found to have statistically significant lower test error performance compared to those selected based on minimum test error. Further results show that the Bayesian Framework is able to identify irrelevant features through Automatic Relevance Determination, though not resulting in a statistically significant error reduction in predictive performance in one-hour ahead forecasting. In day-ahead forecasting removing irrelevant features based on Automatic Relevance Determination is found to yield statistically significant improvements in test error.

Artificial Neural Networks (ANNs) have received increasing attention in recent years with applications that span a wide range of disciplines including vital domains such as medicine, network security and autonomous transportation. However, neural network architectures are becoming increasingly complex and with an increasing need to obtain real-time results from such models, it has become pivotal to use parallelization as a mechanism for speeding up network training and deployment. In this work we propose an implementation of Network Parallel Training through Cannon’s Algorithm for matrix multiplication. We show that increasing the number of processes speeds up training until the point where process communication costs become prohibitive; this point varies by network complexity. We also show through empirical efficiency calculations that the speedup obtained is superlinear.