*2.2. Smart Districts*

The terminology "smart" has in recent years been extensively used and elaborated on. There are several definitions, when it comes to the labelling of smart buildings or smart districts. Wiggington and Harris conclude that there exist more than 30 separate definitions on the term intelligence in relation to buildings [23]. Other literature on the subject states that intelligent buildings are clearly multi-faceted and whilst they can be summarized by a series of characteristics that include aspects ranging from user safety and comfort to resources, a universal description is challenging [24]. Whilst smartness has been mainly used as a label for buildings and cities, the intermediate scale of the district gains in importance as local energy solutions emerge.

Within the contact of this approach, the smartness mostly related to the e fficient use of resources and energy. From an architectural perspective, the planning of a building is mostly limited and defined by the actual plot of the construction. Nevertheless, urban planning considerations as well as scale, morphology and societal setting amongs<sup>t</sup> others heavily influence the design. Energy e fficiency is also largely dependent on the immediate context relating to climate, resources, and infrastructure. Incorporating a systemic view beyond the building's edge towards the district can provide added value, as concepts for sustainable neighborhoods play a fundamental role in the development of resilient cities [17].

Following the logic of resource and energy e fficient design on a buildings scale [25], the district offers the benefit of the systemic perspective and can subsequently deliver optimization beyond the single entity. Especially when it comes to building renovation, clustering buildings with di fferent thermal qualities and connected energy generation, supply and storage systems can achieve significant primary energy savings with minimal physical intervention compared to the renovation of single entities alone [26].

The steps as outlined in Figure 1 describe the methodology for the development of energy concepts at district scale. In step 1, the passive aspects, defined by the architecture (shape, form, envelope, mass) are the first key measures to reduce the actual energy demand for heating, cooling, lighting and ventilation. At the district scale the influencing factors include the orientation of the building blocks, the density and the functional mix of the various entities. The 2nd step relates to the energy systems and focus on the energy networks, the use of waste heat potential and the e fficient control managemen<sup>t</sup> across buildings. In step 3 the adequate selection of RES and respective energy storage solutions is key to move towards zero-carbon district solutions. The overall load managemen<sup>t</sup> as well as small scale district heating or cooling grids are relevant aspects in step 4, where the overall district in connection to the larger urban entity or region signifies the move towards smart district and subsequently smart city solutions.

**Figure 1.** Methodology for resource and energy e fficient design at district scale, author's graphic adapted from [25].

There is however still the question how the domains of the smart buildings can be connected to the district and subsequently to the city to ensure interoperability, especially when it comes to digital planning tools but also for operational purposes. In a recent publication where the interoperability of smart buildings into smart city platforms was evaluated, the authors concluded that the five aspects of smart energy, smart mobility, smart life, smart environment and smart data were the key domains related to the interconnectivity between the scales. Subsequently, smart building integration into a smart city has been defined to set out the framework for the various integration levels [27]. On a building scale there are already several assessments in place that are aimed at quantifying the intelligence or smartness of a building, several of which have been analyzed within the context of the named publication. The building intelligent quotient (BIQ) program developed by the BIQ Consortium consisting of CABA (Continental Automated Building Association) members, is a program aimed at evaluating building intelligence. Whilst it is mainly focused on building automation and control, it should function by its own definition as an evaluation tool for a buildings' smartness [28]. The Honeywell Smart Building Score (HSBS) provides a rating on 15 technology asset groups that make a building green, safe, and productive, and similarly o ffers a broad approach on the rating of devices, software, and control mechanisms within a building [29]. Both ratings encompass a rather complex and elaborate process and mainly focus on intelligent control mechanism and appliances rather than on load management, which provides an entirely di fferent approach as outlined in this paper. In addition, they are also focused on a technology and device-oriented approach, which would need to be adapted as new technologies emerge.

Whilst the assessment of energy flows in buildings is already mostly considered standard practice with more or less detail, the district scale requires a di fferent set of models as other aspects, such as mobility, networks and other infrastructure (e.g., waste heat from industrial processes) need to be factored in. In order to identify optimal strategies at the district level, methodologies should include qualitative and quantitative evaluation procedures based on reciprocal impacts [30]. However, there is also a need for district and energy models that go beyond the scientific community to be applied in practical design developments. Focusing especially on the planning community, the CityCalc tool has been developed to provide a quick assessment for urban planning competitions and initial planning phases [31,32]. Spatial-temporal modeling and thus dynamic assessments on a district scale can also be carried out by the CEA (city energy analyst) [33], a free open-source GIS-(geographic information systems) integrated system that has been conceived as an urban building simulation platform for the analysis and scenario comparison of energy demand, associated CO2 emissions, financial benefits and production optimization for districts [34,35]. GIS based data analysis coupled with energy workflow modelling can be of particular importance for integrated urban platforms, that aim at modelling a diverse range of CO2 emission related domains such as energy, resources and mobility. Such multi-domain tools can help to identify, e.g., high impact districts in need of modernization within the di fferent urban sectors [36].

In a recent study, several urban energy-planning tools have been assessed based on their overall user friendliness related to spatial scale, output time and energy services with a few having been considered suitable for widespread use [37]. District energy modeling also supports the development of adequate technical and economical solutions for the existing building stock, as energy e fficiency and renewable energy measures can be potentially more cost-e ffectively integrated at a multi-building scale. IEA Task 75 is specifically dedicated to the development of solutions for existing urban districts [38]. These developments show, that there is a noticeable shift towards the system perspective, which becomes ever more important, as with the exponential increase in information and communication technology, buildings act as distributed consumers, producers and storage of energy and thus develop into active players in the energy system. Similarly, the clustering of buildings to larger entities becomes more relevant as synergies related to energy e fficiency and renewable energy sources can be exploited [14,15].

In his recently published book on the Green New Deal, the economist Jeremy Rifkin argues that a factor in his so called third industrial revolution towards a de-carbonized energy system lies in the digitalization of the energy networks. He stipulates that paradigm changes can occur when new communication technology converges with new energy sources and new forms of mobility. Thus he concludes that the world is on verge of a third industrial revolution as the Internet is connected to the energy system and to the mass transport systems of e-mobility [39]. Consequently, the assessment of the flexibility of buildings is significant in this context.

#### *2.3. The Potential of Load Shifting in Buildings for the Integration of RES*

The transition towards a sustainable energy system relies heavily both on the lowering of the overall demand and the provision of the required energy by renewable sources. The transformation from a centralized market to an intelligent smart grid requires a fundamental change in how we conceive the production, distribution, storage, and supply of energy [40]. As outlined above, in this context, buildings play a crucial role as they are significant consumers of energy but can at the same time provide surface areas for the integration of de-centralized solar energy systems and storage potential by means of their thermal mass and building services systems.

Aggregating buildings for cooperative energy managemen<sup>t</sup> can yield substantial energy savings by exploiting their load shifting capabilities and utilizing shared energy systems. Aggregating buildings to clusters allow the exploitation of the variation in energy demand in di fferent building types [15]. Determining e fficient control strategies to allow a data driven and robust optimization strategy are necessary to use the potentials at a larger scale [41]. For electrical smart grids (SGs) the integration of renewable energy (RE) generation also depends largely on e fficient demand response (DR). Increasing the share of RES implies that both storage systems and DR have to be jointly considered as with an increasingly higher share of RES the flexibility in the grids decrease without adequate managemen<sup>t</sup> of demand and supply. Studies undertaken in specific micro-grids analyzing the e ffects of high renewable energy penetration highlight that adequate methods must be applied for an e ffective demand response managemen<sup>t</sup> [42]. In order to provide de-centralized storage devices in buildings that increase the participation of end-users in the operation of the grids, micro-storage solutions must be properly planned and managed in order to provide optimized results [43]. Peer to peer (P2P) energy trading can also enable direct energy trading between energy consumers and prosumers. This reduces the exchange between the microgrid and the large-scale utility grids and can subsequently support the wide-ranging penetration of renewable energy into the power grid [44].

Whilst the electrical load shifting undoubtedly dominates the discussions related to smart grids, the thermal integration and consideration of intelligent thermal grids must not be neglected. Integrating PV (photovoltaic) systems in buildings, the so-called building integrated PVs (BIPVs) can, from an architectural and building technical point of view, more easily be achieved compared to solar thermal systems. There is nevertheless still a demand to also incorporate thermal renewables into the urban fabric. Solar thermal collectors, heat pumps, systems based on biomass or waste heat from auxiliary sources can all provide low emission alternatives to fossil-based systems. In a forecast scenario for the European heating and cooling fuel deployment an increase in the share of renewable energy from 16.7% in 2012 to 25.9% in 2030 is possible under the current policy scenario. This is driven mostly by an increased deployment of RES and a simultaneously falling final energy demand due to stricter building e fficiency [45].

There is a growing awareness, that district heating networks should also react to the de-centralization of the energy market and subsequently allow the integration of small-scale supply, mostly based on RES, into their systems [46]. This would also allow the use of so-called waste heat (usually low temperature heat) from industrial processes, wastewater or reject heat from cooling systems. Several studies sugges<sup>t</sup> that there is a significant potential to exploit these ye<sup>t</sup> untapped resources [47,48]. Especially data centers, with their large demand in power and cooling energy represent both a potential for waste heat as well as renewable energy integration. A recent study has found that regional climate studies can provide an e ffective way of improving the e fficiency of data centers in both the upstream renewable energy supply and the downstream waste heat reuse [48].

Although supply temperature from so called prosumers (customers that consume as well as produce energy) is usually lower than typical supply temperature, the thermal networks need to e ffectively manage and control their system to increase the share of decentralized renewable integration [49]. A thorough analysis on the exact scale and potential of the renewable input is however crucial to determine the feasibility of the de-centralized option. Defining a model that combines prosumers, central supply as well as market and emissions aspects can be accomplished by applying stochastic optimization algorithms. However, achieving a fair distribution of economic benefits between a central heat plant and multiple consumers remains a challenging task [50].

A study comparing several scenarios from the (classical) central heat-plant setup, to an agent-based approach and prosumer centric solutions comes to the conclusions that no approach has emerged as superior to the others and that each solution is justified under certain circumstances. It stresses subsequently that mathematical optimization is crucial in determining the best way forward [51]. While economic benefits are achieved in most scenarios, it is a non-trivial task to construct a market model that distributes these benefits in a fair way between the central heat plant and the prosumers.

Focusing on exergy with the aim to use energy e fficiently and reduce carbon emissions presents ye<sup>t</sup> another modelling approach. By considering the match between the grade of energy on the demand and supply side analytical models can provide useful decision support for the planning of low- or zero energy districts [52]. Combining heating and power models by providing modeling solutions for the design of co-generation is an essential cornerstone on the development of decision support mechanisms at the district scale. Multi-criteria optimization allows the focus not just on minimization of operation and maintenance system costs, but also taking into account time-varying loads, tariffs, and ambient conditions [53]. The intelligent coupling of heat and power demand and supply and subsequent co-generation is of particular importance on that scale as thermal and electrical loads can more efficiently be balanced on multiple and different building types with varying demands. Integrating renewable energy systems thus requires a multi-objective approach that considers both economical as well as environmental functions [54]. Quantifying relevant characteristics regarding the generation, distribution, and storage of energy in districts consequently represents a highly relevant aspect in the increased integration of RES into the urban environment. Current assessments and simulation tools on a district scale address the energy related aspects, however load shifting is still mostly considered from the perspective of the utility provider and thus mostly neglected in appraisals focusing on the characteristics of smart buildings and districts.
