Prosumer planning in the DEZENT context of regenerative power production

The electricity is a vital asset and a priority for the social and economic development of today’s world. Building energy infrastructures with high efficiency and renewable energy sources is an important yet challenging task for a sustainable future. Smart grid is a term referring to a modernized electrical grid that uses information and communications technology to gather and act on information, such as information about the behaviors of suppliers and consumers, in an automated fashion to improve the efficiency, reliability, economics, and sustainability of the production and distribution of electricity. The main issue of this thesis is to propose new solutions which help end-users to optimize their consumption and better manage their own electricity costs. More specifically, the challenge is to make elastic the demand for, and the supply of, electricity of prosumers in order to optimize their energy cost based on power market conditions and on suitable constraints on their power consumption. By definition, a prosumer is a user that not only consumes, but also produces and stores electricity. In our work, we focus on power market models in which prosumers interact in a distributed environment during the purchase or sale of electric power. We have chosen to follow the distributed power market model DEZENT. Our contribution is the planning phase of the consumption of prosumers based on the negotiation mechanism of DEZENT. We propose a controller for the planning of the consumption which aims at minimizing the electricity cost achieved at the end of a day. Our controller model exploits the standard dynamic programming algorithm and in the thesis we discuss the assumptions on which the controller design is based. In order to evaluate the performance of our introduced controller, we performed extended experimental studies based on the available DEZENT simulator and on the Java implementation of the optimal controller. Our main result is that the highest energy cost reduction was obtained when we have a high variance on the profile cost of the electricity, the prosumer environment is in the undersupply situation and the reserve capacity of the controller is infinite. Vice versa, the lowest energy cost reduction corresponds to low variance, oversupply situation and finite reserve capacity. Furthermore the study of the problem of (sub) optimal repeated re-planning for the rest of the day has shown that a prosumer having to consume more than expected will pay a remarkable additional cost at the end of the day which depends also on the increased unit costs; in the case in which a part of the available energy reserve is lost, the additional cost paid is proportional to the amount of energy lost. In summary the general idea behind the planning (optimization) phase of a consumer is to plan consumption as smartly (delay or anticipate the consumption) as possible during a day, a week, a month or even a year. The results of this behavior is the minimization of the electricity cost in the long run, under certain assumptions on energy costs, as resulted by local negotiations, and on the acceptable variations of consumer requests. One of the open issues is the global effect of our introduced controller in a prosumer population since each prosumer can make use of it. We believe that the issue can lead to a congestion problem similar to that of the minority game problem proposed in economics literature