Control and management in smart grids are facing many challenges such as scalability, heterogeneity and technology innovation. This requires a transformation from the traditional centralised paradigm into a distributed one. In this article, a new distributed programming methodology, called Centralised Programming and Distributed Execution (CPDE), is proposed. CPDE relies on (i) the abstraction of the whole system as a distributed database; (ii) the use of the Smartlog declarative and reactive rule based language for expressing data manipulation; and (iii) the automatic Smartlog rule distribution according to data distribution. It thus provides a simple and straightforward mean for distributed programming. A centralised algorithm of fair over-voltage regulation of PV systems is used as a typical smart grids study case to validate the methodology and to compare it with centralized implementations. The experiments are implemented in a real-time simulation platform with a network of Raspberry Pis. In addition to showing its correctnes and ease of use, the performance of the CPDE implementation is studied, as well as its sensitivity to the increasing number of computing units and the data distribution. Results are promising and show the clear benefits of this methodology compared to more classical implementations.

Load prediction is at the core of many smart grid topics because its accuracy affects the resilience and optimal operation of the power grid. Short term load forecasting (STLF) specific to buildings is needed for both demand and production sides management, to help reducing peak electricity demand, optimizing energy planning, management, and conservation among other. In this paper, we analyze and compare five state of art regressive STLF algorithms relying on deep learning techniques. Algorithms were chosen based on their performances in the context of load prediction. These methods are compared in the context of smart buildings with real (large) data-set, leading to a comparison on quality metrics and a precise evaluation. The load forecasting is deployed with and without taking into account environmental features to better evaluate the benefits of reaching for external information. The idea is to help determine which of the five forecasting algorithm could be considered in the optimization loop of an energy management systems, either at the level of a district (local energy communities, or smart districts) or directly at the level of individual households (smart residential buildings) using quality metrics and need of exogenous data. Results show that the algorithms will perform better if we accept a larger training time and also present a high sensitivity on the choice of influencing parameters.

In Smart grid ecosystems, it is important to carefully choose the placement of the datasets across different kind of big data systems in order to achieve high performance of the workloads and conformity with the business and data ecosystem. Our approach for datasets placement is based on metadata about datasets, workloads, and systems. This paper gives a general overview of the data placement module, proposes a high-level design and data model for our solution and presents the placement criteria.

In the control and management of smart grids, from steady state to real-time, the objective is to handle and to treat any change in the system as fast as possible, with as less resources as possible. In this context, this paper proposes a new language, called Smartlog, designed as declarative programming. Smartlog is developed for distributed computing in real-time and distributed database management. Compared to imperative programming, based on anticipation rather than reaction, the interest is to not ignore the meaning of some data, or to collect and analyze data without interest, and thus loose bandwidth and computational time. Smartlog is designed for operating smart grids, which are defined as abstract structures of large and scalable distributed databases. After its definition, the main features of the Smartlog language are its compactness, its simplicity and its scalability are shown. The language is tested on the application of a frequency and voltage secondary control of an islanded micro-grid in an experimental test-case, using a realtime simulator connected to Raspberry Pis. The characteristics of Smartlog are illustrated thanks to a comparison with an imperative programming implementation of the same regulation.

An UBIQUEST system provides a high level programming abstraction for rapid prototyping of heterogeneous and distributed applications in a dynamic environment. Such a system is perceived as a distributed database and the applications interact through declarative queries including declarative networking programs (e.g. routing) and/or specific data-oriented distributed algorithms (e.g. distributed join). Case-Based Reasoning is used for optimization of distributed queries when as there is no prior knowledge on data (sources) in networking applications, and certainly no related metadata such as data statistics.

This paper concerns the integration of the Case Based Reasoning (CBR) paradigm in query processing, providing a way to optimize queries when there is no prior knowledge on queried data sources and certainly no related metadata such as data statistics. Our Query Optimization by Learning (QOL) approach optimizes queries using cases generated from the evaluation of similar past queries. A query case comprises: (i) the query, (ii) the query plan and (iii) the measures (computational resources consumed) of the query plan. The work also concerns the way the CBR process interacts with the query plan generation process. This process uses classical heuristics and makes decisions randomly (e.g. when there is no statistics for join ordering and selection of algorithms, routing protocols); It also (re)uses cases (existing query plans) for similar queries parts, improving the query optimization and evaluation efficiency.

This paper focuses on the software architecture and components for supporting core adaptive and interactive query processing. The architecture permits the reuse of most of the existing adaptation techniques and facilitates the development of new ones. Developers will use it to program adaptable and interactive query evaluators, called Query Brokers. Three adaptability levels are considered: (i) static adaptability, achieved at query broker building time, (ii) personalization, achieved for each (user) query having specific needs, and (iii) dynamic adaptability, achieved during query evaluation.