Abstract
Short-term load forecasting plays an important role to the operation of electric systems, as a key parameter for planning maintenances and to support the decision making process on the purchase and sale of electric power. A particular case in this respect is the consumption forecasting on special days, which can be a complex task as it presents unusual load behavior, when compared to regular working days. Moreover, its reduced number of samples makes it hard to properly train and validate more complex and nonlinear prediction algorithms. This paper tackles this problem by proposing a new approach to improve the accuracy of the predictions amidst existing special days, employing an Information Theoretic Learning Mean Shift algorithm for pattern discovery, classifying and densifying the available scarce consumption data. The paper describes how this methodology was applied to an electrical load forecasting problem in the northern region of Brazil, improving the previously obtained accuracy held by the power company.

Abstract
This paper presents a model for breaker status identification and power system topology estimation based on a mosaic of local auto-associative neural networks. The approach extracts information from values of the analog electric variables and allows the recovery of missing sensor signals or the correction of erroneous data about breaker status. The results are confirmed by extensive tests conducted on an IEEE benchmark network.

Abstract
This work presents a methodology for selecting the most relevant measurements for real-time power system monitoring. A genetic algorithm is employed to find the meter plan, composed of relevant, real-time measurements and pseudo-measurements that present the best compromise between investment costs and state estimation performance. This is achieved by minimizing both the number of real-time measurements in the power network and the degradation of the estimated states. Performance measures based on the Information Theory are investigated. Simulation results illustrate the performance of the proposed method.

Abstract
The deployment of Smart Grid technologies opens new opportunities to develop new forecasting and optimization techniques. The growth of solar power penetration in distribution grids imposes the use of solar power forecasts as inputs in advanced grid management functions. This paper proposes a new forecasting algorithm for 6 h ahead based on the vector autoregression framework, which combines distributed time series information collected by the Smart Grid infrastructure. Probabilistic forecasts are generated for the residential solar photovoltaic (PV) and secondary substation levels. The test case consists of 44 micro-generation units and 10 secondary substations from the Smart Grid pilot in Evora, Portugal. The benchmark model is the well-known autoregressive forecasting method (univariate approach). The average improvement in terms of root mean square error (point forecast evaluation) and continuous ranking probability score (probabilistic forecast evaluation) for the first 3 lead-times was between 8% and 12%, and between 1.4% and 5.9%, respectively. (C) 2015 Published by Elsevier Ltd.

Abstract
The growing penetration of solar power technology at low voltage (LV) level introduces new challenges in the distribution grid operation. Across the world, Distribution System Operators (DSO) are implementing the Smart Grid concept and one key function, in this new paradigm, is solar power forecasting. This paper presents a new forecasting framework, based on vector autoregression theory, that combines spatial-temporal data collected by smart meters and distribution transformer controllers to produce six-hour-ahead forecasts at the residential solar photovoltaic (PV) and secondary substation (i.e., MV/LV substation) levels. This framework has been tested for 44 micro-generation units and 10 secondary substations from the Smart Grid pilot in Evora, Portugal (one demonstration site of the EU Project SuSTAINABLE). A comparison was made with the well-known Autoregressive forecasting Model (AR - univariate model) leading to an improvement between 8% and 12% for the first 3 lead-times.

Abstract
The Unit Commitment (UC) problem consists on the day-ahead scheduling of thermal generation units. The scheduling process is based on a forecast for the demand, which adds uncertainty to the decision of starting or shutting down units. With the increasing penetration of renewable energy sources, namely wind power, the level of uncertainty is such that deterministic UC approaches that rely uniquely on point forecasts are no longer appropriate. The UC approach reported in this paper considers a stochastic formulation and includes constraints for the technical limits of thermal generation units, like ramp-rates and minimum and maximum power output, and also for the power flow equations by integrating the DC model in the optimization process. The objective is to assess the ability of the stochastic UC approach to decrease the expected value of load shedding and wind power loss when compared to the deterministic UC approach. A case study based on IEEE-RTS 79 system, which has 24 buses and 32 thermal generation units, for two different penetrations of wind power and a 24-hour horizon is carried out. The computational performance of the methodology proposed is also discussed to show that considerable performance gains without compromising the robustness of the stochastic UC approach can be achieved.