“20th PSCC 2018 papers submission and review platform

Full Program »

Multi-objective probabilistic power resources planning for microgrids with ancillary services capacity

Microgrids have become an attractive strategy to guaranty a reliable, efficient and sustainable electricity supply for remote areas and centers with critical loads. Furthermore, it has been identified that this type of active distribution network (ADN) can positively impact the energy market through the ancillary services that it could offer to the power utility. Nonetheless, the favorable implementation of these microgrids comprises a wide range of engineering challenges, such as those related to the planning and design processes. For instance, the planning methodology have to be reviewed to consider features such as a high penetration level of distributed generation (DG) of intermittent nature, the grid connected and islanded operation modes, active loads with a probabilistic hourly load profile, the use of distributed storage (DS) resources, and the microgrid's capacity to supply ancillary services to the main grid. Therefore, this problem leads to the question about how these particular features can be considered as part of an expansion planning methodology to achieve the most promising characteristics of a microgrid with offer of ancillary services.

In this way, the microgrid planning gives rise to a complex optimization problem, where more than one objective function, several discrete and continuous decision variables, and complicated constraint functions are involved. Additionally, the microgrid mathematical model must consider not only a high penetration level of dispatchable DGs but also non-dispatchable DGs, which are related to renewable generation that leads to a stochastic modeling and a dynamic DS. Furthermore, the penetration of distributed energy resources (DERs) is constrained by the physical area availability and primary energy fluctuation, which normally are not included as part of an unique optimal planning problem.

Therefore, the purpose of this paper is to propose a probabilistic multi-objective microgrid planning methodology based on the optimal DGs' size and location for the minimization of the annual energy losses and maximization of the DG penetration level for the backup power supply as ancillary service. Constraints such as the vacant area and primary energy capacity are considered, and the mathematical model of the microgrid is defined to consider probabilistically more than one geographical primary-energy-availability along the planning horizon.

The methodology is defined to consider three different type of DG technologies: wind turbines, PV modules, and biomass units. Moreover, the probabilistic behavior of their primary energy is included by the Weibull and Log-normal probabilistic distribution functions (PDF), which are attained from both historical data and Monte Carlo Simulation. Thus, the renewable generation output power and load are modeled as multi-state variables.

The test system PG\&E 69-Bus, and state-of-the-art load model and measured primary energy time series were chosen to test the methodology, and the optimization problem is solved using the optimization method Nondominated Sorting Genetic Algorithm II (NSGAII)

The results show that a multi-objective formulation will let to find a set of possible solutions where, for example, the annual energy losses can be reduced up to about 138,82MWh/year, while the DGs supplies 20GWh/year, and the microgrid transfers around 149MWh/year with a penetration level of 60\% of DGs. In short, the main contribution of the full paper is the planning methodology that considers the reverse power that the microgrid can optimally supply to the main grid as an ancillary service together with other objectives, as well as, the multi-state and probabilistic model of the renewable DGs and loads and the inclusion of location constraints.


Sergio Felipe Contreras    
Universidad Nacional de Colombia

Camilo Andres Cortes    
Universidad Nacional de Colombia

Johanna M.A Myrzik    
Institute of Energy Systems, Energy Efficiency and Energy Economics TU Dortmund University


Powered by OpenConf®
Copyright ©2002-2014 Zakon Group LLC