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.

Vulnerability and adaptation to climate change in Guam – a case-study of the perceptions, attitudes, behaviors, and knowledge of a small rural community towards their watershed

This thesis argues that examining the attitudes, perceptions, behaviors, and knowledge of a community towards their specific watershed can reveal their social vulnerability to climate change. Understanding and incorporating these elements of the human dimension in coastal zone management will lead to efficient and effective strategies that safeguard the natural resources for the benefit of the community. By having healthy natural resources, ecological and community resilience to climate change will increase, thus decreasing vulnerability. In the Pacific Ocean, climate and SLR are strongly modulated by the El Niño Southern Oscillation. SLR is three times the global average in the Western Pacific Ocean (Merrifield and Maltrud 2011; Merrifield 2011). Changes in annual rainfall in the Western North Pacific sub‐region from 1950-2010 show that islands in the east are getting much less than in the past, while the islands in the west are getting slightly more rainfall (Keener et al. 2013). For Guam, a small island owned by the United States and located in the Western Pacific Ocean, these factors mean that SLR is higher than any other place in the world and will most likely see increased precipitation. Knowing this, the social vulnerability may be examined. Thus, a case-study of the community residing in the Manell and Geus watersheds was conducted on the island of Guam. Measuring their perceptions, attitudes, knowledge, and behaviors should bring to light their vulnerability to climate change. In order to accomplish this, a household survey was administered from July through August 2010. Approximately 350 surveys were analysed using SPSS. To supplement this quantitative data, informal interviews were conducted with the elders of the community to glean traditional ecological knowledge about perceived climate change. A GIS analysis was conducted to understand the physical geography of the Manell and Geus watersheds. This information about the human dimension is valuable to CZM managers. It may be incorporated into strategic watershed plans, to better administer the natural resources within the coastal zone. The research conducted in this thesis is the basis of a recent watershed management plan for the Guam Coastal Management Program (see King 2014).