Photovoltaic (PV) panels have become a significant source of electric power generation. These panels are considered to be one of the cleanest energy production systems available, so their spread is expected to increase in the following years, especially because recent technologies have reduced the cost of these panels. Unlike classic energy production methodologies that are connected to the high-voltage transmission power lines, many PV panels are connected directly to the lower voltage distribution networks of the electric power grid, making the management of the grid an ongoing challenge. In this letter, we address this challenge and show that the irradiance field that is required to calculate the expected power output of the PV panels can be estimated in a simplistic methodology, using partial observability of the solar radiation. We validate our proposed methodology by conducting an empirical study that uses real data of the solar radiation taken from satellites, and we show that even when the observability of the solar radiation is as low as 10% (meaning that only one in ten points of interest in a regular grid is observable), the irradiance field can be accurately estimated.

In this paper, a mathematical relationship between temperature and solar irradiance is established in order to reduce the sample space and provide joint probabilistic forecasts. These forecasts can then be used for the purpose of stochastic optimization in power systems. A Volterra system type of model is derived to characterize the dependence of temperature on solar irradiance. A dataset from NOAA weather station in California is used to validate the fit of the model. Using the model, probabilistic forecasts of both temperature and irradiance are provided and the performance of the forecasting technique highlights the efficacy of the proposed approach. Results are indicative of the fact that the underlying correlation between temperature and irradiance is well captured and will therefore be useful to produce future scenarios of temperature and irradiance while approximating the underlying sample space appropriately.

The existing approaches to control electrical grids combine frequency and voltage controls at different time-scales. When applied in microgrids with stochastic distributed generation, grid quality of service problems may occur, such as under- or overvoltages as well as congestion of lines and transformers. The COMMELEC framework proposes to solve this compelling issue by performing explicit control of power flows with two novel strategies: 1) a common abstract model is used by resources to advertise their state in real time to a grid agent; and 2) subsystems can be aggregated into virtual devices that hide their internal complexity in order to ensure scalability. While the framework has already been published in the literature, in this paper, we present the first experimental validation of a practicable explicit power-flow primary control applied in a real-scale test-bed microgrid. We demonstrate how an explicit power-flow control solves the active and reactive power sharing problem in real time, easily allowing the microgrid to be dispatchable in real time (i.e., it is able to participate in energy markets) and capable of providing frequency support, while always maintaining quality of service.

We consider line outages in the transmission network of the power grid, and specifically those caused by natural disasters or large-scale physical attacks. In such networks, an outage of a line may lead to overload on other lines, thereby leading to their outage. Such a cascade may have devastating effects not only on the power grid but also on the interconnected communication networks. We study a model of such failures and show that it differs from other models used to analyze cascades (e.g., epidemic/percolation-based models). Inspired by methods developed for network-survivability analysis, we show how to identify the most vulnerable locations in the network. We also perform extensive numerical experiments with real grid data to estimate the effects of geographically correlated outages and briefly discuss mitigation methods. The developed techniques can indicate potential locations for grid monitoring, and hence, will have impact on the deployment of the smart-grid networking infrastructure.

No-regret algorithms has played a key role in on-line learning and prediction problems. In this paper, we focus on a generalized no-regret problem with vector-valued rewards, defined in terms of a desired reward set of the agent. For each mixed action q of the opponent, the agent has a set R(q) where the average reward should reside. In addition, the agent has a response mixed action p which brings the expected reward under these two actions, r(p, q), to R(q). If a strategy of the agent ensures that the average reward converges to R(\barq_n), where \barq_n is the empirical distribution of the opponent’s actions, for any strategy of the opponent, we say that it is a no-regret strategy with respect to R(q). The standard no-regret problem is obtained as a special case for scalar rewards and R(q) = r ∈R: r ≥r(q), where r(q) = \max_p r(p, q). For this problem, the multifunction R(q) is convex, and no-regret strategies can be devised. Our main interest in this paper is in cases where this convexity property does not hold. The best that can be guaranteed in general then is the convergence of the average reward to R^c(\barq_n), the convex hull of R(\barq_n). However, as the game unfolds, it may turn out that the opponent’s choices of actions are limited in some way. If these restrictions were known in advance, the agent could possibly ensure convergence of the average reward to some desired subset of R^c(\barq_n), or even approach R(\barq_n) itself. We formulate appropriate goals for opportunistic no-regret strategies, in the sense that they may exploit such limitations on the opponent’s action sequence in an on-line manner, without knowing them beforehand. As the main technical tool, we propose a class of approachability algorithms that rely on a calibrated forecast of the opponent’s actions, which are opportunistic in the above mentioned sense. As an application, we consider the online no-regret problem with average cost constraints, introduced in Mannor, Tsitsiklis, and Yu (2009), where the best-response-in-hindsight is not generally attainable, but only its convex relaxation. Our proposed algorithm applied to this problem does attain the best-response-in-hindsight if the opponent’s play happens to be stationary (either in terms of its mixed actions, or the empirical frequencies of its pure actions).

We focus on resource allocation for energy harvesting devices. We analytically and numerically evaluate the performance of algorithms that determine time fair energy allocation in systems with predictable and stochastic energy inputs. To gain insight into the performance of networks of devices, we obtain results for the simple cases of a single node and a link. Due to the need for low complexity algorithms, we focus on simple policies (some of which proposed in the past as heuristics) and analytically derive performance guarantees. We also evaluate the performance via simulation, using real-world energy traces that we collected for over a year, and in a testbed of energy harvesting devices developed within the EnHANTs project.

We consider the online binary classification problem, where we are given m classifiers. At each stage, the classifiers map the input to the probability that the input belongs to the positive class. An online classification meta-algorithm is an algorithm that combines the outputs of the classifiers in order to attain a certain goal, without having prior knowledge on the form and statistics of the input, and without prior knowledge on the performance of the given classifiers. In this paper, we use sensitivity and specificity as the performance metrics of the meta-algorithm. In particular, our goal is to design an algorithm which satisfies the following two properties (asymptotically): (i) its average false positive rate (fp-rate) is under some given threshold, and (ii) its average true positive rate (tp-rate) is not worse than the tp-rate of the best convex combination of the m given classifiers that satisfies fp-rate constraint, in hindsight. We show that this problem is in fact a special case of the regret minimization problem with constraints, and therefore the above goal is not attainable. Hence, we pose a relaxed goal and propose a corresponding practical online learning meta-algorithm that attains it. In the case of two classifiers, we show that this algorithm takes a very simple form. To our best knowledge, this is the first algorithm that addresses the problem of the average tp-rate maximization under average fp-rate constraints in the online setting.