*2.2. Methodologies Regarding Fog Based Architecture*

In [21], authors propose a cloud based DSM system using MGs which manages energy for the consumers from multiple regions in order to minimize the utility and consumers' cost. This technique also reduces the time and efforts by incorporating the modularity feature for developing smart cities. Bi-level optimization algorithm using linear cost function is also developed and applied in this scheme. Authors in [22] propose energy management scheme for electricity and natural gas network using integrated DSM. This scheme differentiates the electricity consumption techniques because they consider the consumption regarding each user by considering multiple users' interactions. Nash Equilibrium (NE) is used in this technique for measuring the interactions of the players. NE is applied for electricity cost and peak load reduction.

Authors present a new SG architecture for electric vehicles' scheduling in buildings (integrated with MGs) which enable a central SG controller for cloud Data Center (DC) environment [23]. This system helps the SG users to optimize the load requests in an efficient fashion. It enhances the stability of grid during the high demand intervals. Cloud computing helps the consumers by providing: computing, storage and networking capacities. In addition, there are also some limitations of cloud, i.e., latency, security and downtime problems. In order to ameliorate the latency, reliability, and resiliency of the services, one new strategy is presented in [24], in which fog platform is utilized by considering the energy management as a service. This strategy also discusses two energy management prototypes: home energy management prototype and MG energy management prototype for electricity bill and delay minimization. Security for the customers' private data is enhanced through encryption, which improves the flexibility and reliability by augmenting the fog computing principles [25,26]. It is also responsible for providing the locality of consumers' appliances.

Afterwards, a new approach regarding electricity cost reduction problem presents the internet payload requests and SG dynamic electricity pricing mechanism [27]. This approach also discusses the predictive cost control for the smart charging on both sides: (1) battery energy for the servers and (2) electricity from the power grid. In order to mitigate the overall cost of the system, batteries are charged during the low price rate hours and are discharged during the high price rate hours. In [28], authors use big data analytics for balancing the energy of the residential, industrial and commercial buildings. Multiple big data anaytics and decision making functions are used for efficiently managing the energy in these buildings. This work also considers the comfort preferences of the consumers using the control configuration and planning processes. This technique is designed for the buildings of smart cities.

The aforementioned schemes are described for the load scheduling of appliances in the buildings. However, integrated cloud and fog based scheme is not described for optimized resource allocation in any of the regions of the world. The proposed work describes the energy management in the residential buildings using C2F2C framework.
