Sebastian Kochanneck

Organic Architecture for Energy Management and Smart Grids

An unprecedented rise of renewable and distributed energy resources imposes unprecedented challenges in terms of complexity to power grids. Multitudes of devices are not only connected to the electricity grid but need appropriate information and communication technologies for proving their services. These devices ask for novel control mechanisms on different levels and regional scales. In this paper, we show how concepts from Organic Computing may support the controlled self-organization of the future smart grid. We propose a generic hierarchical architecture as a framework for various energy management systems. This architecture is able to reflect the physical power grid structure as well as the interdependencies of its stakeholders, user objectives, subsystems, and devices. It enables adaptive responses to changing objectives as well as disturbances in the system. Various simulations of systems based on the proposed architecture show the applicability of the proposed architecture to domains of energy management in smart grids.

Detecting Occupancy in Smart Buildings by Data Fusion from Low-cost Sensors: Poster Description

This poster presents ongoing work on an occupancy detection system based on data fusion of various low-cost sensors within a real smart building and compares the results of different classification approaches.

A Microservice Architecture for the Intranet of Things and Energy in Smart Buildings: Research Paper

This paper presents challenges and issues in smart buildings and the Internet of Things (IoT), which we identified in years of research in real buildings. To tackle these challenges, a decentralized service-oriented architecture based on a message-oriented middleware has been implemented for the domain of smart buildings. It uses a network-transparent IoT message bus and provides the means for composing applications from auxiliary services, which facilitate device abstraction, protocol adaption, modularity, and maintainability. We demonstrate the flexibility of our architecture by describing how three distinct applications---privacy-enhancing energy data visualization, automated building energy management, and a generic user interface---can be integrated and operated simultaneously in our real smart building laboratory. We compare the advantages of our architecture to conventional ones and provide a best-practice solution for the Intranet of Things and Energy in smart buildings.

A Privacy-Aware Architecture for Energy Management Systems in Smart Grids

Electrical power grids are in a phase of transition. Therefore, control mechanisms are needed that allow the integration of renewable energies, decentralized generation and storage systems, as well as electric vehicles into the arising smart grid. In this paper, we propose a privacy-aware architecture based on the principles of Organic Computing to tackle the aforementioned challenges. It enables controlled self-organization on all levels, from in-house optimization through demand side management to mains utility and management of entire smart grids. Physical as well as virtual devices and entities can be incorporated flexibly into the control loop of the proposed architecture. Tofu fill privacy requirements, a component named Data Custodian Service is introduced. It handles data exchange across system borders while protecting privacy and interests of data owners. Privacy is preserved by giving the data owner the choice to control data access based on temporal and spatial resolution and on precision of the requested data sets. The application to real world scenarios is shown using the example of a building energy management system that implements the suggested architecture.

Reference Scenarios for Low Voltage Power Systems: Poster Description

This poster presents three electric reference scenarios for a rural area, a village, and a suburb, providing data for the grids, the connected households as well as distributed generation. The scenarios resemble the situation in Germany, giving data for the present as well as future parametrizations and making it available to a larger public for the evaluation of smart grid optimization and control algorithms.