ISGT 2016 Tutorials will be given on Sunday, October 9, 2016. To register, please visit the Registration portal. Information on the tutorials and instructors is shown below.


Demand Response Management in Smart Grid
Speaker: Yan Zhang, Simula Research Laboratory & University of Oslo, Norway

Description: Smart power grid is a modernized grid that proactively uses state-of-the-art technology in the areas of sensing, communications, control, computing, and information technology to improve efficiency, sustainability, stability, and security. Demand response management (DRM) plays a critical role in balancing and shaping the electricity demand and supply. The electricity providers aim to schedule power generation according to the power demand obeying the physical constraints of the power system. The consumers reshape their demand profiles in response to the supply conditions. Both sides are rational and thereby aim to maximize their own profits. With the development of information technologies, power technologies and smart transportation, the smart grid is facing unprecedented challenges. These include distributed architecture, renewable energy resources, energy storage, and widespread electric vehicles. These new features make DRM a significant challenge.
In this tutorial, we will introduce demand response management basic concepts, main approaches and recent advances. Different pricing schemes with examples will be discussed. In addition, we will also introduce our recent studies on intelligent demand response solutions for the control and optimization of smart grid with different approaches (e.g., game theory and queuing theory). Finally, we will present DRM applications in reducing energy consumption in data centers.


Load Forecasting with Artificial Intelligence on Big Data
Speakers: Patrick Oliver GLAUNER, University of Luxembourg; Radu STATE, University of Luxembourg

Description: In the domain of electrical power grids, there is a particular interest in time series analysis using artificial intelligence. Machine learning is the branch of artificial intelligence giving computers the ability to learn patterns from data without being explicitly programmed. Deep Learning is a set of cutting-edge machine learning algorithms that are inspired by how the human brain works. It allows to self-learn feature hierarchies from the data rather than modeling hand-crafted features. It has proven to significantly improve performance in challenging signal processing problems. In this tutorial, we will first provide an introduction to the theoretical foundations of neural networks and Deep Learning.
Second, we will demonstrate how to use Deep Learning for load forecasting with TensorFlow, Google’s in-house Deep Learning platform made for Big Data machine learning applications. The advantage of Deep Learning is that the results can easily be applied to other problems, such as detection of nontechnical losses. Attendees will be provided with code snippets that they can easily amend in order to perform analyses on their own time series.


 

Smart Energy Technologies for Anomaly Detection in Safeguards Applications
Speakers: Miltiadis Alamaniotis, Purdue University, USA; Lefteri H. Tsoukalas, Purdue University, USA

Description: This tutorial will present preliminary efforts and challenges imposed in developing smart energy technologies utilized for anomaly detection and identification in safeguards applications. As energy shifts from demand-led to supply-constrained, coupling power systems with information systems has converted traditional electric energy delivery infrastructures into an interconnected large-scale hybrid energy-data system. In addition, technological and economical advancements in information collection and storage, power systems, electricity grid, and communication technologies are driving a transition from old traditional structures to fully technologically equipped structures identified as smart buildings. Smart buildings, either residential or industrial or commercial, are fully instrumented with a large variety of sensors and smart devices installed in them. To this end, smart electrical appliances and loads in smart buildings are connected to an intelligent meter that acquires operational data from each appliance and performs tasks pertaining to electricity purchase and consumption management. The high data acquisition rates together with the increasing number of smart appliances imply that a large volume of heterogeneous data is gathered at the intelligent meter, and is being transmitted over the smart power grid.
Furthermore, technological advancements in machine learning, data mining and big data computing, introduce new opportunities regarding monitoring and surveillance technologies that are applicable in developing next generation systems for enhancing safeguards. In particular, consumption patterns together with the plethora of information signals transmitted via the intertwined energy and communication networks may be processed and used to make inferences about anomalies that exceed normal grid operation; such anomalies may result from an event that consists of a threat to public and global safety. For instance, acquired power grid data may be used to detect an undeclared critical facility, which is well hidden in an urban environment. At a more advanced stage, energy data may be used to identify the specific operation of the facility and/ or verify its role with respect to its local environment.
The focus of this tutorial will be on exploring the capabilities of smart energy technologies whose deployment allows detection of anomalies pertained to electricity consumption of interest to safeguards. To this end, we will present the challenges in safeguards and present novel machine intelligence tools for signal processing and knowledge extraction in the context of smart energy systems. An overview of existing detection tools regarding smart power systems will be presented and implications for future research activities and challenges will be discussed. Overall, for the smart power system community, the tutorial will provide an approach for utilizing the power grid data in the aspects of modeling, simulation, signal processing and computational efficiency pertaining to safeguards and security.
The proposed tutorial has not been presented before and will include work that is at its first stages.


Distributed and hierarchical congestion management in distribution network containing distributed energy resources
Speakers: Sami Repo, Tampere University of Technology; Anna Kulmala, Tampere University of Technology

Description: Congestion management is one of the key components in the transfer towards active distribution networks that enable cost-effective network interconnection of distributed generation and better utilization of network assets. Although congestion management has been a subject of active research in the last years only few real network implementations exist. Demonstrations are vital to enable large scale utilization of congestion management algorithms. At present, distribution system operators usually resort to network reinforcement in case of foreseen congestions. This approach, however, leads to high network total costs in many cases. The proposed tutorial will present the distributed automation architecture and hierarchical congestion management concept developed and implemented in IDE4L project [1]. Demonstration results both from laboratory and from field will be shown. The proposal is for a two time slots tutorial (3 hours).
The distributed automation architecture developed in IDE4L project [2] enables efficient control of the whole distribution network utilizing all available controllable resources, also those connected to LV networks. Substation automation units (SAUs) added to primary and secondary substations take care of monitoring and control of one MV or LV network and only necessary data is transferred between the different level SAUs and the control center. The distributed architecture decreases the amount of data transfer in the network, simplifies connecting new distributed energy resources to the control system and makes the architecture more robust against component failures. The control architecture utilizes standard data models (IEC 61850 and CIM model). The tutorial will present the proposed control architecture in detail. Its mapping to SGAM framework is also discussed.
The congestion management hierarchy [3] consists of three levels: primary controllers operate based on local measurements, secondary control optimizes the set points of the primary controllers in real-time and tertiary control utilizes load and production forecasts as its inputs and realizes network reconfiguration algorithm and connection to the market place. Primary controllers are located at the connection point of the controllable resource, secondary controllers at primary and secondary substation SAUs and tertiary control at the control center. Hence, the control is spatially distributed and operates in different time frames. The tutorial will discuss the different hierarchical levels and the interactions between different actors in detail.
Demonstration results both from laboratory and from field will be presented. The laboratory demonstrations utilize Real Time Digital Simulator (RTDS) simulation environment. Real control devices such as generator voltage controllers and a real SAU implementation are connected to the distribution network emulated by the RTDS. Field demonstrations are conducted in three different locations: Denmark (Østkraft Holding A/S), Italy (A2A Reti Electtriche SpA) and Spain (Unión Fenosa Distribución, S.A.). Results from the demonstrations will be presented in the tutorial. The operation of the automation architecture and control concept will be evaluated.


Techno-economic modeling and multi-service optimization of smart buildings, smart districts and community-based microgrids
Speakers: Dr. Eduardo Alejandro Martínez Ceseña, The University of Manchester; Dr. Nicholas Good, The University of Manchester; Dr. Pierluigi Mancarella, The University of Manchester

Description: The emergence of Distributed Energy Resources (DER) coupled with Information and Communications Technologies (ICT) is providing end-users with unprecedented flexibility to mitigate economic and environmental costs in a Smart Grid context. This has motivated the rise of smart buildings, which can exploit multi-energy vectors (e.g., electricity, heat and gas) to flexibly meet their energy needs and sell services to other actors (e.g., aggregators, retailers, network operators etc.), to support efficient energy system operation. For example, a smart building may have the flexibility to meet its heating needs using gas (e.g., with a gas boiler) or electricity (e.g., with an electric heat pump) based on the economic costs and potentially carbon intensity of each energy vector. In addition, buildings can exploit various forms of storage flexibility, such as thermal storage inherent in the building fabric, which needs to be assessed against users’ loss of comfort, or such as electrical storage via electric vehicles.
This flexibility and efficiency can be further increased by coordinating the operation of smart buildings at the district level for optimal energy management of clusters of buildings or district energy systems (for instance based on decentralized combined heat and power), with potential to provide multiple services such as reserves and local network capacity support. Special cases of such smart (multi-energy) districts are represented by community-based microgrids that could operate in both islanded and grid-connected mode, with further reliability and economic benefits.
However, modeling and optimization of single/multiple buildings within a smart district are daunting tasks calling for an explicit understanding of: (i) the energy services required by the building occupants; (ii) technical characteristics and constraints of the buildings, local and district-level DER (including different types of storage), and different energy networks (e.g., electricity, heat, gas) that interconnect the buildings; and (iii) the economic implications of multiple services that could be provided throughout the value chain to multiple actors.
This tutorial provides a holistic overview of technical and economic modeling and optimization principles of smart buildings, districts and community-based microgrids based on latest research at the University of Manchester in various UK and European projects, including:

  • High-resolution stochastic energy service modeling of residential buildings equipped with heat pumps, solar photovoltaic, electric vehicles, etc;
  • Different forms of flexibility that can be provided by smart buildings and districts, including from various types of thermal and electrical storage;
  • Demand response assessment, including quantification of economic benefits and impact on end-user utility/comfort, of smart building and districts;
  • Optimal energy management of district energy systems, smart districts; and islanded and grid-connected community-based microgrids, considering multi-energy network load flows and constraints;
  • Optimal provision of multiple (e.g., energy/reserve/capacity/reliability) services from smart districts;
  • Cost benefit analysis and business case assessment of multi-energy district operation under different traditional and smart operational regimes.

Real world applications will be used to demonstrate the above concepts and in particular how the flexibility inherent in smart buildings and districts can mitigate costs and emissions and boost their business case by providing a wide range of services in the energy chain.


Advanced Modelling of Smart Distribution Networks Using OpenDSS
Speaker: Prof Luis (Nando) Ochoa, The University of Melbourne and The University of Manchester

Description: The increasing and future adoption of small-to-medium scale low carbon technologies such as wind power, photovoltaic systems and electric vehicles is and will pose significant technical and economic challenges on distribution networks. Medium and low voltage circuits have been designed to have no or limited controllability and hence are largely unmonitored. However, it is likely that they will become one of the first bottlenecks towards the decarbonisation of our power systems. Therefore, it is important to understand the impacts and the potential solutions in the context of Smart Grids. For this purpose, it is crucial to use simulation tools designed specifically for distribution networks and flexible enough to carry out sophisticated studies.
This 3-hour tutorial will give the attendees the opportunity to learn about the basic and advanced applications of OpenDSS, an open source state-of-the-art distribution network analysis software package developed by EPRI (USA), in the context of Smart Distribution Networks. The tutorial includes hands on aspects for a direct familiarization with OpenDSS as well as details of the modelling frameworks needed to produce more advanced studies. To illustrate this three industrial Smart Grid projects in the UK are presented considering the interactions with other analysis and optimisation software packages (e.g., Matlab and AIMMS).


IEC 61850 and its role in the Smart Grid
Speaker: Dr. Alexander Apostolov, PAC World

The tutorial introduces the concept of the Smart Grid with its main characteristics, components and communications interfaces. This is followed by a description of the object modeling principles of IEC 61850 and the services available to meet the requirements of different applications. The use of IEC 61850 for engineering of Smart Grids based on the System Configuration Language (SCL) defined in the standard is later presented. The quality assurance process defined in the standard is also introduced. The concept of fully digital substations together with the expansion of IEC 61850 beyond the substation are presented at the end.


IEEE 1547 – Standard for Interconnecting Distributed Energy Resources with Electric Power Systems
Speaker: Dr. Babak Enayati, National Grid

This tutorial will introduce the IEEE 1547 “Standard for Interconnecting Distributed Energy Resources with Electric Power Systems”.
Due to the increasing amount of Distributed Energy Resources (DERs) interconnections with the Electric Power System, the IEEE 1547 standard is going through a major revision to address some of the technical issues associated with high penetration of DERs. The participants will learn about the status of the standard revision and the recent proposed draft changes to the standard i. e. voltage regulation, response to abnormal system conditions (including voltage and frequency ride through), power quality, etc.
The participants will also learn about the utility concerns/solutions to adopt the revised IEEE 1547 standard.


IEEE Std. 998 – DIRECT LIGHTNING STROKE SHIELDING OF SUBSTATIONS
Speaker: James “Brian” Cain, PE, IAEI

This 4-hour tutorial will provide a review of the recently released IEEE Std. 998 guide. Empirical and EGM shielding methods for direct lightning stroke interception (including the newly added Eriksson EGM) will be reviewed and examples presented in step-by-step detail. The basic principles behind four alternative models for direct lightning stroke interception will be presented and discussed with a practical example.

Tutorial Outline:
• Overview of changes between 1996 and 2012 version
• Lightning characteristics
• Lightning exposure and substation risk assessment
• Review of traditional methods with examples: fixed angle and empirical curves
• Review of basic electrogeometric model with examples: rolling sphere method
• Review of the improved EGM with examples
• Description of four alternative models for substation shielding: LPM, CVM, LIT and SLIM
• Example shielding designs with the four alternative models
• Conclusions
• Q & A