Avdevicius, Edvard

The emergence of new energy optimisation and control technologies with the concept of power system flexibility is a promising way to achieve the desired optimum, secure management within the smart grid and green energy transition. In this context, demand response is available through flexible demand management, taking into account various technical and time constraints. Accordingly, the aim of this paper is to address existing constraints in the field of electric mobility, in particular the operation of the charging infrastructure of bus depots, in order to actively and effectively participate in demand response events by forecasting day-ahead charging costs and load profiles of public transport infrastructure. In line with the development of a methodology for forecasting more accurately, this paper develops a prediction model based on machine learning (ML). A charging schedule is then produced based on Mixed-Integer Linear Programming (MILP) with various objective function scenarios, taking into consideration the electricity price forecast and load distribution. As a result, calculating new provisional load profiles involves assessing the flexibility potential of the bus fleet and preparing solutions in advance based on the electricity market situation.

In recent years, electric vehicles (EVs) are considered to be a promising way to reduce greenhouse gas emissions from the transportation sector. However, the increasing penetration of EVs into the distribution network (DN) raises serious concerns about the network’s safe and reliable operation. The uncontrolled EV charging with random behavior will lead to volatile load peaks on the distribution transformer. In order to obtain more transformer loading capacity available for integration of further EVs, distributed energy resources (DERs) and related devices, such as heat pumps, the transformer loading must be limited to a certain range. For this reason, an intelligent charging management based on model-free Reinforcement Learning (RL) is proposed in this work. The RL management is able to control the charging power of all EVs connected to the network without previous knowledge about the arriving- and leaving time. The needed information for the RL-agent to perceive the current state of the system is formed with cumulated values such as the total energy requirement and the total charging power demand of all EVs. In this paper, the RL algorithm is trained on real-world energy consumption data for a month and on a reference network, created with selected characteristics of a substation network area in the northeast of Hamburg. Comparing with uncontrolled charging, the simulation results show that the RL-based charging management avoids 99 % of threshold violations regarding transformer loading and results in 1% of EV energy requirement is not satisfied. Through sensitivity analysis regarding the state space representation in the employed RL process, the necessity of providing the state of charge (SOC) or the energy requirements of EV users are proven to improve the charging control performance.