Gabriel de Oliveira Ramos

Dynamic constrained coalition formation among electric vehicles

BackgroundThe use of electric vehicles (EVs) and vehicle-to-grid (V2G) technologies have been advocated as an efficient way to reduce the intermittency of renewable energy sources in smart grids. However, operating on V2G sessions in a cost-effective way is not a trivial task for EVs. The formation of coalitions among EVs has been proposed to tackle this problem.MethodsIn this paper we introduce Dynamic Constrained Coalition Formation (DCCF), which is a distributed heuristic-based method for constrained coalition structure generation (CSG) in dynamic environments. In our approach, coalitions are formed observing constraints imposed by the grid. To this end, EV agents negotiate the formation of feasible coalitions among themselves.ResultsBased on experiments, we show that DCCF is efficient to provide good solutions in a fast way. DCCF provides solutions whose quality approaches 98% of the optimum. In dynamically changing scenarios, DCCF also shows good results, keeping the agents payoff stable along time.ConclusionsEssentially, DCCF’s main advantage over traditional CSG algorithms is that its computational effort is very lower. On the other hand, unlike traditional algorithms, DCCF is suitable only for constraint-based problems.

Self-Adapting Coalition Formation Among Electric Vehicles in Smart Grids

In the last years, the need for using multiple energy sources made the concept of smart grids emerge. A smart grid is a fully automated electricity network, which monitors and controls all its elements being able to supply energy in an efficient and reliable way. Within this context, the use of electric vehicles (EVs) and Vehicle-To-Grid (V2G) technologies have been advocated as an efficient way to reduce the intermittent supply associated with renewable energy sources. However, operating on V2G sessions in a cost effective way is not a trivial task for EVs. To address this problem, the formation of coalitions among EVs has been proposed as a mean to improve profitability on V2G sessions. Addressing these scenarios, in this paper we introduce the Self-Adapting Coalition Formation (SACF) method, which is a local and dynamic heuristic-based mechanism for coalition structure generation. In our approach, coalitions are formed observing constraints imposed by the grid to the EVs, which negotiate locally the formation of feasible coalitions among themselves. Based on experiments, we see that SACF is an efficient method, providing good solutions in a simple and low-cost way. SACF is faster than centralized methods and provides solutions with near optimal quality. In dynamic scenarios, SACF also shows very good results, being able to keep the agents gain relatively stable along time, even in quickly changing environments. Its main advantage is that the computational effort is very low, while classical centralized methods are limited to manage no more than a dozen agents in a reasonable amount of time.

Forming Coalitions of Electric Vehicles in Constrained Scenarios

Finding an optimal coalition structure is a hard problem. In order to simplify this process, it is possible to explore some characteristics of the agents organization. In this paper we propose an algorithm that deals with a particular family of games in characteristic function, but is able to search in a much smaller space by considering organizational issues such as constraints in the number of participants. We apply this approach to the domain of smart grids, in which the aim is to form coalitions of electric vehicles in order to increase their reliability when supplying energy to the grid.

A self-adapting similarity-based coalition formation approach for plug-in electric vehicles in smart grids

The Vehicle-To-Grid (V2G) concept is a key feature towards the integration of electric vehicles into smart grids. Through V2G sessions, plug-in electric vehicles (PEVs) can sell their surplus energy to the grid. However, profiting from V2G sessions is not trivial for singletons. Thereby, the formation of coalitions among PEVs has been proposed to tackle this issue. In this paper, we rely on domain knowledge in order to propose a novel modelling for such problem. Specifically, we investigate how the similarity among the PEVs’ energy profiles can be used to improve the formation of coalitions. The energy profile of a PEV estimates how long such PEV will be available for the V2G session. Based on this, we aim at maximising the coalitions’ duration. We empirically show that our approach is both efficient (it outperforms state-of-the-art algorithms in terms of runtime) and effective (solutions were 96.5% of optimum, on average).

Towards the User Equilibrium in Traffic Assignment Using GRASP with Path Relinking

Solving the traffic assignment problem (TAP) is an important step towards an efficient usage of the traffic infrastructure. A fundamental assignment model is the so-called User Equilibrium (UE), which may turn into a complex optimisation problem. In this paper, we present the use of the GRASP metaheuristic to approximate the UE of the TAP. A path relinking mechanism is also employed to promote a higher coverage of the search space. Moreover, we propose a novel performance evaluation function, which measures the number of vehicles that have an incentive to deviate from the routes to which they were assigned. Through experiments, we show that our approach outperforms classical algorithms, providing solutions that are, on average, significantly closer to the UE. Furthermore, when compared to classical methods, the fairness achieved by our assignments is considerably better. These results indicate that our approach is efficient and robust, producing reasonably stable assignments.

Coalitions of Electric Vehicles in Smart Grids

In this chapter, we introduce the use of self-organised coalitions in smart grid scenarios for finding a coalition structure that maximises the systems’ utility. The complexity of such a task is exponential with the number of agents, and optimal coalition formation has been considered impractical. Several heuristic alternatives have been proposed in the research literature to handle such a problem. However, most existing methods approach coalition formation neglecting important aspects like maximising the total revenue or ensuring stability. Nonetheless, these points are fundamental in the context of smart grids, especially when we refer to virtual power plants (VPPs) of plug-in electric vehicles (PEVs), which have very limited energy capacity and small profits. In this chapter, we present two classes of constraints: (i) geographic-based, where the geographic position of PEVs is considered to avoid overloading the energy distribution network; and (ii) user-based, where the preferences of the PEV-users (owners) are taken into account to promote lasting coalitions. We also propose three methods for addressing coalition formation within such constrained scenarios: (i) DCCF, where agents invite neighbours to join their coalitions; (ii) SACF, where agents ask to join their neighbours’ coalitions; and (iii) SACF\(^+\), which is a natural evolution of SACF, where agents can change their coalitions, thus making the process much more dynamic. In all cases, agents negotiate the formation of coalitions among themselves, each on behalf a single PEV. The presented approaches were evaluated in closed and open world scenarios. Regarding the results, all three methods run in a few milliseconds regardless of the number of agents, achieving near-optimal solutions. In all tested cases, results were above 90% of optimum, on average. In comparison, despite delivering optimal solutions, traditional approaches took several hours and run for up to 20 agents, which represents a small and unrealistic scenario for smart grids. Thus, the proposed approaches show that providing approximate solutions for the coalition formation problem is attainable in smart grids scenarios.

Towards a Platform for Testing and Developing Privacy-Preserving Data Mining Applications for Smart Grids

In this paper we analyse the trade-off between privacy-preservation methods and the quality of data mining applications, within the specific context of the smart grid. The use of smart meters to automate data collection is set to solve the problem of electricity theft, which is a serious concern in developing nations. Nevertheless, the unlimited use of data from smart meters allows for potentially private information to be discovered. There is a demand for methods to quantify the trade-off between privacy-preservation and quality of a classification model. We describe the research and development of an agent-based simulation platform to evaluate the balance between privacy-preservation mechanisms and methods for electricity theft detection. We have implemented a proof-of-concept model and validated it against real data collected from smart meters.