Power system parameters forecasting using Hilbert-Huang transform and machine learning

A novel hybrid data-driven approach is developed for forecasting power system parameters with the goal of increasing the efficiency of short-term forecasting studies for non-stationary time-series. The proposed approach is based on mode decomposition and a feature analysis of initial retrospective data using the Hilbert-Huang transform and machine learning algorithms. The random forests and gradient boosting trees learning techniques were examined. The decision tree techniques were used to rank the importance of variables employed in the forecasting models. The Mean Decrease Gini index is employed as an impurity function. The resulting hybrid forecasting models employ the radial basis function neural network and support vector regression. A part from introduction and references the paper is organized as follows. The second section presents the background and the review of several approaches for short-term forecasting of power system parameters. In the third section a hybrid machine learningbased algorithm using Hilbert-Huang transform is developed for short-term forecasting of power system parameters. Fourth section describes the decision tree learning algorithms used for the issue of variables importance. Finally in section six the experimental results in the following electric power problems are presented: active power flow forecasting, electricity price forecasting and for the wind speed and direction forecasting.

Toxicity of manufactured nano-ZnO to Lemna minor

The rapid development of nanotechnology has led to a rise in the large-scale production and commercial use of engineered nano-ZnO. Engineered/manufactured nano-ZnO are applied in a broad range of products such as drugs, paints, cosmetics, abrasive agents and insulators. This can result in the unintended exposure of human beings to nano-ZnO and will inevitably result in the release of nano-ZnO in to the environment. Thus, it is necessary to assess the risk of nano-ZnO to the environment. In this thesis the toxicity of nano-ZnO was analysed using the aquatic, primary producer lesser duckweed (Lemna minor), and the mechanism of toxicity was analysed. Both short-term (one week) and long-term (six weeks) toxicity of nano-ZnO (uncoated) were determined. Results show that the toxicity of nano-ZnO added to the aquatic growth medium increases with increasing concentration and that toxicity accumulates with exposure time. A study of nano-ZnO dissolution reveals that the main reason for nano-ZnO toxicity on Lemna minor is the release of Zn ions. Nano-ZnO dissolution is pH dependent, and toxicity matches the release of Zn2+. Functional coating materials are commonly added to nano-ZnO particles to improve specific industrial applications. To test if coating materials contribute to nano-ZnO toxicity on lesser duckweed, the effect of silane coupling agent (KH550) coated nano-ZnO on Lemma minor was investigated. Results show that coating can decrease the release of Zn ions, which reduces toxicity to Lemna minor, in contrast to uncoated particles. Another commonly hypothesized reason for nano-ZnO toxicity is the formation of Reactive Oxygen Species (ROS) on the particles surface. As part of this thesis, the ROS formation induced by nano-ZnO was studied. Results show that nano-ZnO catalyse ROS formation and this can negatively affect duckweed growth. In conclusion, this work has detailed potentially toxic effects of nano-ZnO on Lemna minor. This study has also provides references for future research, and informs regulatory testing for nanoparticle toxicity. Specifically, the outcomes of this study emphasize the importance of exposure time, environmental parameters and coating material when analysing NPs toxicity. Firstly, impacts of longer exposure time should be studied. Secondly, environmental parameters such as pH and medium-composition need to be considered when investigating NPs toxicity. Lastly, coating of NPs should always be considered in the context of NPs toxicity, and similar NPs with different coatings require separate toxicity tests.