Abstract
The motivation for this work is the study and prediction of wind ramp events occurring in a large-scale wind farm located in the US Midwest. In this paper we introduce the SHRED framework, a stream-based model that continuously learns a discrete HMM model from wind power and wind speed measurements. We use a supervised learning algorithm to learn HMM parameters from discretized data, where ramp events are HMM states and discretized wind speed data are HMM observations. The discretization of the historical data is obtained by running the SAX algorithm over the first order variations in the original signal. SHRED updates the HMM using the most recent historical data and includes a forgetting mechanism to model natural time dependence in wind patterns. To forecast ramp events we use recent wind speed forecasts and the Viterbi algorithm, that incrementally finds the most probable ramp event to occur. We compare SHRED framework against Persistence baseline in predicting ramp events occurring in short-time horizons, ranging from 30 minutes to 90 minutes. SHRED consistently exhibits more accurate and cost-effective results than the baseline. © 2012 Springer-Verlag Berlin Heidelberg.

Abstract
This paper presents a sequential Monte Carlo simulation algorithm that can simultaneously assess composite system adequacy and detect wind power curtailment events. A simple procedure at the end of the state evaluation stage is proposed to categorize wind power curtailment events according to their cause. Furthermore, the dual variables of the DC optimal power flow procedure are used to identify which transmission circuits are restricting the use of the total wind power available. In the first set of experiments, the composite system adequacy is assessed, incorporating different generation technologies. This is conducted to clarify the usual comparisons made between wind and thermal technologies which, in fact, depend on the performance measure selected. A second set of experiments considering several wind penetration scenarios is also performed to determine the operational rules or system components responsible for the largest amount of wind energy curtailed. The experiments are carried out on configurations of the IEEE-RTS 79 power system.

Abstract
This paper presents a new approach to incipient fault diagnosis in power transformers, based on the results of dissolved gas analysis. A set of autoassociative neural networks or autoencoders is trained, so that each becomes tuned with a particular fault mode or no fault condition. The scarce data available forms clusters that are densified using an Information Theoretic Mean Shift algorithm, allowing all real data to be used in the validation process. Then, a parallel model is built where the autoencoders compete with one another when a new input vector is entered and the closest recognition is taken as the diagnosis sought. A remarkable accuracy of 100% is achieved with this architecture, in a validation data set using all real information available.

Abstract
This paper presents a transformer failure diagnosis system based on Dissolved Gases Analysis that was developed by using a new methodology for extracting fuzzy rules from Kohonen Self-Organizing Map. Firstly, the Kohonen net was trained in order to capture the knowledge from a database of faulty transformers inspected in service. Once the knowledge was captured during the learning stage, it was transformed into the form of Zero-order Takagi-Sugeno fuzzy rules. In the form of fuzzy rules, the relationship between the variables of the system became explicit which have led to a more reliable diagnosis system. Additionally to the extraction of the fuzzy system, a fuzzyfication process was applied in the fuzzy system output. Experimental results demonstrated the efficiency of the diagnosis system proposed that had superior results as compared with other conventional and intelligent methods.

Abstract
This paper discusses the development of a Multi-Agent Systems (MAS) technology-based platform with potential applications in management and simulation processes in power systems. In order to explore some of the features of MAS, a new methodology is proposed to assess power systems reliability based on Monte Carlo simulation (MCS), exploiting the benefits of the distributed artificial intelligence area and, mainly, the use of the distributed capacity in two ways: building autonomous behaviors to the applications and mitigating computational effort. Through the use of this technology, it was possible to divide the MCS algorithm into distinct tasks and submit them to the agents' processing. Two different approaches to solve generating capacity reliability problems based on chronological MCS illustrate the potential of MAS in power systems reliability assessment.

Abstract
This paper proposes a multi-objective approach to a distribution network planning process that deals with the challenges derived from the integration of Distributed Generation (DG). The proposal consists of a multi-objective version of the well-known Evolutionary Particle Swarm Optimization method (MEPSO). A broad performance comparison is made between the MEPSO and other multi-objective optimization meta-heuristics, the Non-dominated Sorting Genetic Algorithm II (NSGA-II) and a Multi-objective Tabu Search (MOTS), using two distribution networks in a given DG penetration scenario. Although the three methods proved to be applicable in distribution system planning, the MEPSO algorithm has presented promising attributes and a constant and high level performance when compared to other methods.