Distributed service infrastructure for monitoring, management and simulation in Smart Cities

World’s energy consumption concentrates within our cities, due to an irreversible urbanisation process. At the same time, insufficient and uncoordinated efforts try to cope with the challenges of urban energy efficiency optimisation. New tailored control policies should be designed for our energy distribution networks. However, to address this task it is necessary to model different entities of our cities (e.g. buildings and energy distribution networks). For instance, in the case of district heating optimisation, the physical model of the district thermal behaviour takes as inputs building energy profiles and signatures, together with information about the district heating network (i.e. network topology and structural attributes). A good description of the actual district entities is instrumental to correctly design and simulate new energy optimisation policies. In addition, novel policies can only be monitored and validated by retrieving information on the actual district energy consumption. This information is extracted by deploying distributed online technologies directly on the eld, to sense and actuate control commands on energy network endpoints. Recently, the introduction of Internet-of-Things technologies generated new business opportunities for the different competitors in the Smart City market. Following the patterns described in the Adjacent Possible theory for innovation, they enabled the deployment of fine-grained monitoring and control policies for public and private spaces (e.g. buildings). The pervasive nature of IoT technologies allows for a new vision of computing, where devices collect data and fade in the background of our environments, and ambient intelligence unifies user awareness on energy consumption and comfort impact of energy optimisation policies. IoT technologies are increasingly adopted because they are versatile and cheap. Thanks to their network capabilities they can be integrated into comprehensive infrastructures. In the last decades, the Smart City community presented different solutions to provide smart environments, mostly used to build energy-aware houses. Some of the challenges which need to be addressed are related to technology interoperability. Indeed, most of the technologies that have been introduced in the market are difficult to integrate due to proprietary communication protocols and data formats. In addition, the extension of current solutions to the district level is not feasible, because they do not consider additional data sources (e.g. Geographical Information Systems) which are needed for the optimal modelling of city districts. Recent literature approaches the definition of an infrastructure for energy management at the district level. Currently, state-of-art does not include such an infrastructure. This work proposes a city district IoT-enabled software infrastructure for energy monitoring, management and simulation. Its purpose is to collect, process and analyse heterogeneous kinds of data. This infrastructure integrates and correlates historical energy consumption with structural features of the different entities of the district (e.g. buildings or energy distribution networks). IoT devices are first-class citizens and they are integrated by using open standards of the Web (i.e. Web Services). The different information models, of the entities which belong to the district, are exposed as a single and consistent District Information Model (DIM). The main challenges addressed by this infrastructure are: • The transparent integration of heterogeneous IoT devices and district level information sources; • The definition of a uniform Web Service-oriented interface across all components of the infrastructure. The contribution of this infrastructure to the state-of-art consists of: • A single platform to integrate and correlate all the different components be- tween each other and with environmental sensors of a city district information model; • A framework for district energy management and optimisation policies simulation. To assess the relevance of the presented infrastructure, two applications which exploit the transparent integration of district information are presented. These two applications retrieve structural and parametric data from the different information models (e.g. geographical maps or building models). Information is then presented to different stakeholders for building benchmarking or energy consumption monitoring of the district. In addition, we designed a case study to test the simulation capabilities of the infrastructure. In this case study, it is depicted how to develop a novel energy peak smoothing policy for a district heating network, and how to validate it both at the district and at the building level. At the district level, it is possible to estimate the reduction of primary energy usage for the energy provider, while at building level the simulation framework assesses the comfort impact for the building’s inhabitants