Complexities arising inside and around emerging electricity systems prompt a multi-layered approach in which different disciplines and areas of expertise are pooled together.
Emerging electricity systems needs to be studied and understood as complex socio-technical systems with multiple physical, cyber, social, policy, and decision making layers; these layers also interact with changing external conditions (economic cycles, technological innovation, and prevailing and changing weather and climatic conditions). Many actors interact within this broader “system of systems”, such as prosumers, distributors, retailers, regulators and policy makers. They act via distributed decision
21
making processes which impact the physically constrained network via diverse electronic means (from control and command systems to smart meters).
The complexity of the future electricity systems (smart grids) is acknowledged as a challenge by CENELEC that has visualized this complexity in the Smart Grid Architecture Model (SGAM) (CEN- CENELEC-ETSI Smart Grid Coordination Group, 2012). In the SGAM approach presented in Figure 5, the Smart Grid Plane covers the complete electrical energy conversion chain that includes generation, transmission, distribution, distributed electrical resources (DER) and customers. The zones represent the hierarchical levels of power system management and the five layers represent business objective and processes, functions, information exchange and models, communication protocols and components. Each layer covers the smart grid plane, which is spanned by electrical domains and information management zones.
Figure 5: SGAM framework (CEN-CENELEC-ETSI Smart Grid Coordination Group, 2012)
The complexity of the smart electricity system rests on the multiplicity of interacting players that operate with, and within, a defined environment as independent decision makers with behaviours that are driven by individual as well as socially driven goals and attitudes. The broader socio-technical network forms a community with high levels of interaction and integration.
While much research has looked at the purpose and functionality of smart electricity systems, smart electricity systems themselves are merely one system in a “system of systems”. As such, complexity is not just an attribute of the smart electricity systems alone, but also the systems interacting with it. For example, the increasing complexity of weather and climate, the increased complexity of social behaviour and the interaction of individuals guided by narrow economic rationality, the complexities of crisis management and emergency response and the overall organisational structure needed to
22
manage all those complexities, must all be studied and modelled to adequately meet the emerging challenges that modern society will face (CEN-CENELEC-ETSI Smart Grid Coordination Group, 2012).
In this context, in order to understand the complexity of future electricity systems, there is the need to move focus and attention from a component-oriented to an interaction-oriented view of the electric power system. The goal of this systemic understanding is to identify tools and techniques for optimal decision-making that will enable society to achieve its energy, environmental, economic and social goals. The framework that should be developed will enable the identification of emerging problems and will provide new solutions and approaches (Bompard et al., 2012).
Complexity sciences can help in modelling and analysing the dynamics and interactions of a broad range of actors and components constituting the technical, social and environmental aspects of smart energy systems thus assisting in investigating present and future challenges in and around future smart energy systems.
A smart energy system includes both local smart distribution grids - characterized by numerous independent participants like prosumers, retailers, distributed-generators, energy storage, EVs as well as technologies still to be invented - and transnational super grids - e.g. connecting large-scale time- varying renewable sources to national power grids and markets).
The main characteristics of these systems are (Bompard et al., 2012):
pervasive deployment of information and communication technologies (ICT);
integration of renewable generation in support of energy, environmental and other policies; bidirectional communication and power exchange between suppliers and
consumers/prosumers;
multiplicity of interacting players operating with, and within, a defined architecture/market; enhanced network flexibility and reliability in a future smart energy system;
newly required approaches for the monitoring, control and protection of power systems in both space and time.
Furthermore technical power systems will operate under varying environmental conditions, exchanging transactions in the power markets. A key concept in complexity science is "emergence". Though some emergent properties can already be anticipated, it is expected that important emergent properties of the future electricity systems remain unforeseen.
The hypothesis is that complexity sciences can help in identifying tools and techniques for optimal decision making encompassing policy and regulatory design, planning and investment, as well as real
23
time operations. Smart energy systems research incorporating complexity sciences can provide models and guidelines for future developments, and for recognizing emerging behaviours and challenges.
As suggested by Bompard (Bompard et al., 2014), a multilayer platform model includes the power, cyber, social and environment layers, along with threats and factors that may affect the system as presented in Figure 6.
Figure 6. Multilayer platform of complex power system (Bompard et al., 2014)
The physical layer includes all the hardware components and contains the electricity flows where physical variables and indices are computed, monitored, or optimized by the system operators. Here, the focus is mainly on MV/LV distribution networks, while including data and constraints from the upper transmission level. The cyber layer is the container of information flows, where all the operations and market-related data sets are managed: prosumer generation and consumption values, market prices, physical conditions, operational commands, and so on. Also various technical innovations are required in this layer, such as smart meters, optical and power-line communications, home area communications, wide area measurement system, and so on. The social layer aggregates the actors of electricity system, i.e. users, prosumers and marketers. This layer has been identified as the main source of complexities that lead to the unpredictable performance of the overall system. The value sense of each prosumer is initially decided by factors related to their psychology, education, profession and so on, and then evolving through the interactions within their social network. Also the
24
social network itself is evolving through random relationships and the establishment of new or interruption of existing social links. The environment layer stands for the natural phenomena, and influences all the other layers. Most obviously, weather conditions, geographical conditions, and primary and secondary resource conditions, directly affect consumption and generation. Society typically imposes sustainability goals that require the respect of several environmental targets (emission, energy efficiency, and so on), with a key role for regulators.
Apart from environmental targets set by society, the overall performance of the system should include all the other dimensions: physical performance in terms of power security, power quality, reaction under emergency; technical performance in terms of technology penetration, technical reliability and efficiency; social performance in terms of satisfaction of the objectives set by regulations, individuals, and social groups; and market performance in terms of market power allocation, competitiveness, etc. (Bompard et al., 2014). Of great importance are the interconnections between the different layers, as these connections are at the basis of the arising complexity. For example the weather conditions impact on the social layer (people behaviour) and on the physical layer (e.g. distributed generation); the physical layer exchanges data with the cyber layer, which performs measurements and provides commands, but influences also the environment layer (e.g. with pollutant emissions); the cyber layer is the mean for prosumers (social layer) to interact with the grid (physical layer). The decision-making processes interact with the other four layers. For example people (social layer) can exercise pressure on politicians for changes in the performance objectives; on the other hand, decision makers can obtain information from the cyber layer and provide commands to it, or can act directly on the physical layer (e.g. the system operators).
As mentioned in section 1.1, the complexity is not only related to the multiplicity of layers, but it is also related to the multi-scale dimension of the energy system itself and to the plurality of the available non-equivalent representations (Kovacic et al., 2015b) that relate to different issues at different level of the system: demand side management, efficiency, reduction of distribution losses, integration of renewable resources and availability of natural resources. The term "electricity grid" has different interpretations depending at which level of the system the analysis is addressed. A representation of the different levels of analysis used to study the performance of electricity grid is proposed in Figure 7 as suggested by Kovacic. Each level of analysis is characterized by a different "system identity" and "potential system use". It is the coexistence of non-equivalent representations of the performance of smart grids that generate ambiguity in the interpretation of what an electric grid is and should be clearly understood and specified.
25
Figure 7. Different level of analysis used to study the performance of electricity grids (Kovacic et al., 2015b)
The scope of this thesis is the analysis and simulation of the interface between the technical system (represented by the smart grid technologies) and the social layer, in the specific the household consumer. I therefore focus on level n-1.