Publication: Modelling of optimal placement and sizing of battery energy storage system using hybrid whale optimization algorithm and artificial immune system for total system losses reduction.
Loading...
Date
2021-06
Authors
Wong Ling Ai
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Battery energy storage system (BESS) is a crucial aspect that supports the operation
of power system, particularly with the increasing penetration of renewable energy (RE)
based generation. Both optimal location and sizing of BESS have become critical
issues in benefitting the advantages offered by BESS in grid. Besides, an optimization
algorithm with high efficiency is important to ensure the attainment of optimal
solutions, where the optimization algorithms like genetic algorithm and particle swarm
optimization are known to have high possibility of being trapped in local optimal
points. Having those said, this study proposes BESS optimisation to reduce the total
system losses using modified Whale Optimisation Algorithm (WOA) with high
exploration and exploitation features. First, two approaches, namely two-step
optimisation approach and simultaneous optimisation approach, were incorporated to
obtain optimal locations and sizing of BESS. The performances of both approaches in
varied scenarios and cases were observed and compared. The approach that exerted
better performance was applied for the next optimisation work, in which optimal
locations and sizing of BESS were determined for distribution network with solar
photovoltaic (PV) and dynamic load profile. Additionally, the duck curve issue (over-generation and steep ramping) served as a constraint in the optimisation process. The
impact of duck curve issue as a constraint on the optimal placement and sizing of BESS
was analysed. A hybrid of WOA and artificial immune system (WOA-AIS) was
initiated to address the constrained optimisation. For all the optimisation work carried
out in this study, the objective function was primarily to minimise the total system
losses. Lastly, the effectiveness of WOA and WOA-AIS in attaining optimal solutions
was validated with other well-known optimisation algorithms, including particle
swarm optimisation (PSO) and firefly algorithm (FA). As a result, the simultaneous
optimisation approach yielded better solutions than the two-step optimisation
approach. Grounded on similar BESS sizing, it was found that the multiple placement
of BESS in distribution network reduced losses more effectively than placing only a
single BESS in the network. Meanwhile, it was found that the BESS energy and total
system losses increased when duck curve issue was considered as constraint in the
optimization process. Upon comparing with other optimisation algorithms, WOA and
WOA-AIS displayed satisfactory performance in obtaining optimal BESS locations
and sizing in distribution network. WOA-AIS demonstrated the improvement of
0.15% and 0.39% than those obtained from PSO-AIS and FA-AIS respectively, for
optimal BESS allocation considering duck curve issue as constraint. This study
concludes that for the application of losses reduction, BESS appears to be most
effective when placed at buses with higher load demand, wherein optimal locations
and sizing of BESS are dictated by the scenario setting and the constraints applied
throughout the optimisation process. The optimal BESS allocation methods proposed
in this thesis not only help in alleviating power system problems, but also solve the
duck curve problem with the increasing penetration of solar PV in the power system.
Battery energy storage system (BESS) is a crucial aspect that supports the operation of power system, particularly with the increasing penetration of renewable energy (RE) based generation. Both optimal location and sizing of BESS have become critical issues in benefitting the advantages offered by BESS in grid. Besides, an optimization algorithm with high efficiency is important to ensure the attainment of optimal solutions, where the optimization algorithms like genetic algorithm and particle swarm optimization are known to have high possibility of being trapped in local optimal points. Having those said, this study proposes BESS optimisation to reduce the total system losses using modified Whale Optimisation Algorithm (WOA) with high exploration and exploitation features. First, two approaches, namely two-step optimisation approach and simultaneous optimisation approach, were incorporated to obtain optimal locations and sizing of BESS. The performances of both approaches in varied scenarios and cases were observed and compared. The approach that exerted better performance was applied for the next optimisation work, in which optimal locations and sizing of BESS were determined for distribution network with solar photovoltaic (PV) and dynamic load profile. Additionally, the duck curve issue (over-generation and steep ramping) served as a constraint in the optimisation process. The impact of duck curve issue as a constraint on the optimal placement and sizing of BESS was analysed. A hybrid of WOA and artificial immune system (WOA-AIS) was initiated to address the constrained optimisation. For all the optimisation work carried out in this study, the objective function was primarily to minimise the total system losses. Lastly, the effectiveness of WOA and WOA-AIS in attaining optimal solutions was validated with other well-known optimisation algorithms, including particle swarm optimisation (PSO) and firefly algorithm (FA). As a result, the simultaneous optimisation approach yielded better solutions than the two-step optimisation approach. Grounded on similar BESS sizing, it was found that the multiple placement of BESS in distribution network reduced losses more effectively than placing only a single BESS in the network. Meanwhile, it was found that the BESS energy and total system losses increased when duck curve issue was considered as constraint in the optimization process. Upon comparing with other optimisation algorithms, WOA and WOA-AIS displayed satisfactory performance in obtaining optimal BESS locations and sizing in distribution network. WOA-AIS demonstrated the improvement of 0.15% and 0.39% than those obtained from PSO-AIS and FA-AIS respectively, for optimal BESS allocation considering duck curve issue as constraint. This study concludes that for the application of losses reduction, BESS appears to be most effective when placed at buses with higher load demand, wherein optimal locations and sizing of BESS are dictated by the scenario setting and the constraints applied throughout the optimisation process. The optimal BESS allocation methods proposed in this thesis not only help in alleviating power system problems, but also solve the duck curve problem with the increasing penetration of solar PV in the power system.
Battery energy storage system (BESS) is a crucial aspect that supports the operation of power system, particularly with the increasing penetration of renewable energy (RE) based generation. Both optimal location and sizing of BESS have become critical issues in benefitting the advantages offered by BESS in grid. Besides, an optimization algorithm with high efficiency is important to ensure the attainment of optimal solutions, where the optimization algorithms like genetic algorithm and particle swarm optimization are known to have high possibility of being trapped in local optimal points. Having those said, this study proposes BESS optimisation to reduce the total system losses using modified Whale Optimisation Algorithm (WOA) with high exploration and exploitation features. First, two approaches, namely two-step optimisation approach and simultaneous optimisation approach, were incorporated to obtain optimal locations and sizing of BESS. The performances of both approaches in varied scenarios and cases were observed and compared. The approach that exerted better performance was applied for the next optimisation work, in which optimal locations and sizing of BESS were determined for distribution network with solar photovoltaic (PV) and dynamic load profile. Additionally, the duck curve issue (over-generation and steep ramping) served as a constraint in the optimisation process. The impact of duck curve issue as a constraint on the optimal placement and sizing of BESS was analysed. A hybrid of WOA and artificial immune system (WOA-AIS) was initiated to address the constrained optimisation. For all the optimisation work carried out in this study, the objective function was primarily to minimise the total system losses. Lastly, the effectiveness of WOA and WOA-AIS in attaining optimal solutions was validated with other well-known optimisation algorithms, including particle swarm optimisation (PSO) and firefly algorithm (FA). As a result, the simultaneous optimisation approach yielded better solutions than the two-step optimisation approach. Grounded on similar BESS sizing, it was found that the multiple placement of BESS in distribution network reduced losses more effectively than placing only a single BESS in the network. Meanwhile, it was found that the BESS energy and total system losses increased when duck curve issue was considered as constraint in the optimization process. Upon comparing with other optimisation algorithms, WOA and WOA-AIS displayed satisfactory performance in obtaining optimal BESS locations and sizing in distribution network. WOA-AIS demonstrated the improvement of 0.15% and 0.39% than those obtained from PSO-AIS and FA-AIS respectively, for optimal BESS allocation considering duck curve issue as constraint. This study concludes that for the application of losses reduction, BESS appears to be most effective when placed at buses with higher load demand, wherein optimal locations and sizing of BESS are dictated by the scenario setting and the constraints applied throughout the optimisation process. The optimal BESS allocation methods proposed in this thesis not only help in alleviating power system problems, but also solve the duck curve problem with the increasing penetration of solar PV in the power system.