Abstract
The ascending need for more energy demand around the globe is a lasting reality. The drive for energy demand is projected to keep accelerating in emerging markets and developing economies. At the same time, the world aims to achieve a global net zero by replacing fossil fuel sources with more environmentally friendly energy sources. Various policies supporting the global net zero emissions goal have contributed to substantially raising the deployment of renewable energy sources (RES), also referred to as distributed generation (DG), as they are installed near energy consumption points. This situation reduced renewables costs, permitting additional RES deployment, further reducing costs, etc. DG grid integration is set to expand in capacity, location, and complexity. Recent data reveal that DG grid integration is sustained by a rapid and significant solar photovoltaic (PV) and wind generation (WG) deployment mainly because these resources do not deplete or emit greenhouse gases and have low operational costs. Extensive PV and WG grid deployment, however, is adversely slowed down by their variability and unpredictability nature, location, and weather dependency, as well as numerous other challenges, prompting power systems (PS) planners and operators to constantly seek innovative and effective ways of mitigating the impact of those operational challenges on particularly stability, reliability, efficiency, and economic viability. The dichotomous energy context described above can be compared to a two-objective problem with antagonistic goals depending upon some uncertainties, multiple constraints, and several dynamic parameters, thus requiring multi-objective-multi-level optimization solution approaches. It is undeniably conflicting to plan to maximize RE grid integration while concurrently attempting to minimize the inherent effects of consequential challenges arising from RE grid integration. Moreover, uncertainties associated with WG, combined with the distribution feeder demand and electricity real-time pricing (RTP) issues, harmfully impact operational limit violations and deepen the financial losses of utility companies. Furthermore, traditional distribution systems (DS) were not designed to handle widespread renewables as high levels of DG penetration pose considerable challenges, such as voltage fluctuation, increased congestion, reverse power flow, demand-supply mismatch, and accelerated equipment aging, to name a few. Consequently, there is an unrelenting need to develop suitable frameworks considering technical and financial implications and efficient strategies to widen RES deployment and increase hosting capacity (HC).
Firstly, this thesis presents a novel bi-layer multi-objective formulation planning model for a solution framework that optimizes a WG grid penetrated network considering the impact of
vi
multiple uncertain stochastic WG, DR program implementation, and battery energy storage system (BESS) incorporation under increased distribution operational challenges. Furthermore, the considered planning model is emphasized for one year while the operational impact of BESS and real-time pricing demand response (RTP-DR) is considered for a short time. The first layer of the bi-level optimization framework is developed for the optimal planning of WGs and BESS capacities while minimizing the costs of energy loss and reverse power flow, respectively, and maintaining node voltage deviation within acceptable limits. The second level is developed to determine the optimal dispatch of BESS and to control flexible loads participating in the RTP-DR program, with the goals of minimizing the cost of energy taken from the grid and broadening HC by increasing WG penetration size. Finally, to handle the complexity of the proposed stochastic planning problem, a bi-level improved crow search optimization algorithm (BICSOA) is formulated and proposed for both levels, considering power balance, feeder current limit, BESS state of charge (SOC), and DR constraints.
Secondly, the thesis developed a novel robust optimization model that presents the best operational planning schedule and tracks the network operational status against consumers' flexible demand while simultaneously determining the capacity of uncertain solar PV scenarios to achieve DS tri-objective operational planning that ensures maximum solar PV hosting capacity, while minimizing voltage deviation and network losses respectively, under the consideration of imperative technical constraints namely: DR, power balance, node voltage limits and solar PV hosting capacity. The modified crow search optimization (MCSO) algorithm is developed and applied to manage the dispatch of the energy system, obtain the optimal scheduling of the proposed energy system, and improve system performance.
The efficacy of the proposed bi-layer framework and its formulated optimization algorithm and tri-objective optimization is satisfactorily tested on a modified standard 33-Bus distribution system. The results evidenced that: (1) when optimally incorporated, BESS dispatch combined with an optimal RTP-DR scheduling of responsive load demand, can efficiently absorb multiple variability uncertainties to achieve a decrease of up to 70 % of annual energy loss, 88 % reduction of annual grid consumption cost, a peak-to-valley distance decrease of up to 74 %, a substantial decrease in feeder current as well as an improvement of the mean node voltage thus widening the HC of WG; (2) with the aid of DR programs, the DS planning and operating model achieved to optimize the HC while minimizing network losses and voltage deviation. These results aligned with the challenge of responding to the growing energy demand and improving accessibility to clean energy.