Abstract
This paper presents a method to identify the status (open or closed) of breakers in network branches, in the absence of status signal or electric measurements on the branch including the breaker. Indirect power measurements from the SCADA are combined to form a 2D image array, which is fed into a Convolutional Neural Network. The image construction is based on ranking measurements with the Cauchy-Schwarz divergence between two signal distributions (for breaker open and closed). The success rate obtained with this technique is close to 100% in the IEEE testbed adopted. © 2019 IEEE.

Abstract
This paper describes an open standards-based information system that can support the democratization and consumer empowerment through flexibility activation in the distribution networks of the near future. The paper outlines a software infrastructure focused on technical issues, closely following the OpenADR standard and the corresponding IEC 62746-10 standard. The infrastructure represents a communication backbone allowing the connection, registering, activation and reporting for different types of granular consumer flexibility. The flexibility sources can be very diverse - from controllable charging set points of electric vehicle chargers and district-level storages such as stationary batteries, towards taking advantage of comparatively large time constants of thermal systems in residential and commercial buildings. In the viewpoint of the proposed system, all these flexibility provisions represent distributed energy resources in a wider sense. The system thus offers interoperable support for consumer-level integration of different energy systems (electricity, heat and gas), and additional flexibility sources are made available to the electric power system, all the time keeping the user comfort and avoiding service disruptions. The paper outlines the technical infrastructure as a backbone activating new sources of flexibility, helping the further proliferation of renewable energy sources and establishing new market actors.

Abstract
Phasor Measurement Units (PMUs) in transmission systems is one of the most promising sources of data to increase situational awareness of network monitoring. However, the inclusion of PMU measurements along with the ones from traditional Supervisory Control and Data Acquisition (SCADA) systems to perform state estimation brings additional challenges, such as the vast difference in sampling rates and precision between these two types of measurements. This paper formally introduces a Bayesian inference approach in the form of a new State Estimator for transmission systems able to deal with the different sampling rates of those measurements. The proposed approach provides accurate state estimates even for buses that are not observable by PMU measurements, and when load variation occurs during the time interval between two SCADA data scans. Several simulation results (with IEEE transmission test systems) are used to illustrate the features of the proposed approach.

Abstract
This work discusses the solution of a Large-scale global optimization problem named Security Constrained Optimal Power Flow (SCOPF) using a method based on Cross Entropy (CE) and Evolutionary Particle Swarm Optimization (EPSO). The obtained solution is compared to the Entropy Enhanced Covariance Matrix Adaptation Evolution Strategy (EE-CMAES) and Shrinking Net Algorithm (SNA). Experiments show the approach reaches competitive results.

Abstract
This paper proposes an estimation strategy in order to address two arising trends in Power System State Estimation (PSSE). Through a hybrid two-stage estimation architecture, high quality measurements gathered by PMUs can be incorporated into PSSE without excluding the widespread employed SCADA measurements. In the first stage of the proposed estimation architecture, SCADA and PMU measurements are individual processed by Maximum Correntropy-based estimators that replace conventional WLS-based methods. The second stage makes use of fusion methods to optimally combine the estimates provided by the individual estimators in order to enhance the quality of final estimates. This architecture allows the inclusion of the new class of measurement while making the whole process bad data-resilient, due to the outlier-rejection properties of Maximum Correntropy-based algorithms.

Abstract