Sofija Nikolova Poceva

Prof. Dr. Sofija Nikolova Poceva

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 August 2026 in Energies
Verified
Journal Article
Open access
PDF file
Pande PopovskiPande Popovski
Pande Popovski
GVGoran Veljanovski
Goran Veljanovski
MAMetodija Atanasovski
Metodija Atanasovski
SPSofija Nikolova Poceva
Sofija Nikolova Poceva
ACAnton Chaushevski
Anton Chaushevski
This paper studies a risk-averse stochastic unit commitment framework for an energy aggregator, operating a portfolio of conventional generators, renewable units, and battery energy storage in a network-constrained environment. Renewable generation and demand uncertainty are represented through a scenario-based extensive-form mixed-integer linear program. To avoid exposure to rare but high cost events, the model incorporates conditional value-at-risk as part of the objective function. The approach captures key market interactions, including day-ahead commitments, imbalance penalties, and power exchange with a neighboring network, while respecting generator constraints, storage dynamics, line flow limits, and bus voltage security. A comprehensive parametric study is conducted to quantify the influence of two risk parameters: the conditional value-at-risk confidence level α and the risk-aversion weight λ. Using a 300-scenario test set on a modified IEEE 9-bus system, the results show that risk-neutral scheduling exposes the aggregator to larger operational costs in extreme scenarios. Minor levels of risk aversion (0.1–0.5) reduce CVaR and tighten the distribution of costs. Increasing λ further yields diminishing returns, while higher α values focus risk mitigation on the most severe outcomes. The results demonstrate how CVaR-based stochastic scheduling can support aggregator decision-making by quantifying downside risk under renewable uncertainty.

13 July 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
Pande PopovskiPande Popovski
Pande Popovski
GVGoran Veljanovski
Goran Veljanovski
MAMetodija Atanasovski
Metodija Atanasovski
SPSofija Nikolova Poceva
Sofija Nikolova Poceva
ACAnton Chaushevski
Anton Chaushevski
This paper studies a risk-averse stochastic unit commitment framework for an energy aggregator, operating a portfolio of conventional generators, renewable units, and battery energy storage in a network-constrained environment. Renewable generation and demand uncertainty are represented through a scenario based extensive-form mixed-integer linear program. To avoid exposure to rare but high cost events, the model incorporates Conditional Value-at-Risk as part of the objective function. The approach captures key market interactions, including day-ahead commitments, imbalance penalties, and power exchange with a neighboring network, while respecting generator constraints, storage dynamics, line flow limits, and bus voltage security. A comprehensive parametric study is conducted to quantify the influence of two risk parameters: the Conditional Value-at-Risk confidence level α and the risk-aversion weight λ. Using a 300-scenario test set on a modified IEEE 9-bus system, the results show that risk-neutral scheduling exposes the aggregator to larger operational costs in extreme scenarios. Minor levels of risk aversion (0.1–0.5) reduce CVaR and tighten the distribution of costs. Increasing λ further yields diminishing returns, while higher α values focus risk mitigation on the most severe outcomes. The results demonstrate how CVaR-based stochastic scheduling can support aggregator decision-making by quantifying downside risk under renewable uncertainty.

1 December 2022 in Energija Ekonomija Ekologija
Verified
Journal Article
Closed access
ACAnton Chaushevski
Anton Chaushevski
SPSofija Nikolova Poceva
Sofija Nikolova Poceva
In the paper alternatives for expansion of the Macedonian electricity supply system are presented. MESSAGE tool is used for modelling the system, for developing the scenarios, to analyse cost optimal energy pathways and to determine the optimal electricity generation technology mix. The calculations for the period 2020-2050 are made in a 5-year interval. Two scenarios are developed: Scenario 1 – BAU (Business as Usual) scenario, where the current energy structure prevailed, with trends like the current development. The electricity demand is satisfied by coal-fired thermal power plants, gas-fired thermal power plants, hydro power plants and renewable power plants. Scenario 2 – Green scenario, where gas-fired thermal power plants and nuclear power representative with small modular reactors are base load technologies, and also intensive construction of production capacities on renewable energy sources is forced. By making further analysis, the possibility of flexibility in the construction of new facilities within the interval, economic indicators, and appropriate impact on the environment are obtained. In order to make an economic analysis for a particular technology, calculations are made for parameters that are standard for projects, such as: net present value, benefit – cost ratio, payback period and internal rate of return. The environmental impact is analysed of aspect of CO2 emissions.

14 September 2026 in Preprints.org
Verified
Preprint
Open access
PDF file
Pande PopovskiPande Popovski
Pande Popovski
GVGoran Veljanovski
Goran Veljanovski
MAMetodija Atanasovski
Metodija Atanasovski
SPSofija Nikolova Poceva
Sofija Nikolova Poceva
ACAnton Chaushevski
Anton Chaushevski
Electricity market settlement rules shape the financial exposure that energy aggregators face when generation deviates from contracted positions, yet their interaction with Conditional-Value-at-Risk-based risk-aversion strategies in aggregator scheduling remains unexplored. This paper studies how the design of imbalance penalty rules influences optimal risk-aversion in a stochastic unit commitment framework for a network-constrained aggregator operating a diversified portfolio of generation resources. Four settlement regimes are defined: a regime with halved deviation penalty rates, a baseline regime matching empirical market data, a strict regime with doubled penalty rates, and a self-balancing regime that prohibits contracted day-ahead exchange. A parametric sweep of the risk aversion weight λ at a fixed confidence level α is conducted on a modified IEEE 9-bus test system using 50 stochastic scenarios. To resolve the transition between the discrete regimes, a finer penalty sweep is also performed. Results show that stricter settlement designs increase tail-risk exposure at the risk-neutral baseline, while the benefit of risk weighting is largest under the strict and self-balancing regimes. Battery energy storage throughput responds to λ under market-participatory regimes but becomes invariant under self-balancing constraints. The findings provide quantitative guidance on how settlement regime should inform aggregator risk management strategy.

13 February 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
Pande PopovskiPande Popovski
Pande Popovski
GVGoran Veljanovski
Goran Veljanovski
MAMetodija Atanasovski
Metodija Atanasovski
SPSofija Nikolova Poceva
Sofija Nikolova Poceva
ACAnton Chaushevski
Anton Chaushevski
This paper studies a risk-averse security constrained stochastic unit commitment framework for an energy aggregator, operating a portfolio of conventional generators, renewable units, and battery energy storage in a network-constrained environment. Renewable generation and demand uncertainty are represented through a scenario based extensive-form mixed-integer linear program. To avoid risk from high costs, the model incorporates Conditional Value-at-Risk as part of the objective function, enabling the aggregator to trade off expected profit against exposure to rare but high-cost events. The approach captures key market interactions, including day-ahead commitments, imbalance penalties, and power exchange with the upstream grid, while respecting generator constraints, storage dynamics, and line flow limits. A comprehensive parametric study is conducted to quantify the influence of two risk parameters: the Conditional Value-at-Risk confidence level α and the risk aversion weight λ. Using a 50-scenario test set on a modified IEEE 9-bus system, the results show that risk-neutral scheduling exposes the aggregator to large profit losses in extreme scenarios, despite achieving high expected profit. Moderate levels of risk aversion (0.2–0.5) reduce Conditional Value-at-Risk and tighten the distribution of profits with negligible impact on expected daily profit. Increasing λ further yields diminishing returns, while higher α values focus risk mitigation on the most severe outcomes. The results demonstrate how CVaR-based stochastic scheduling can support aggregator decision-making by quantifying the tradeoff between expected profit and downside risk under renewable uncertainty.

1 January 2020 in Energija Ekonomija Ekologija
Verified
Journal Article
Closed access
SPSofija Nikolova Poceva
Sofija Nikolova Poceva
ACAnton Chaushevski
Anton Chaushevski
VSValentino Stojkovski
Valentino Stojkovski
ZMZoran Markov
Zoran Markov
The paper presents the methodology for hydro power plants operation (Long-term and Short-term) in a hydro power cascade system. It includes water managing of generating and/or pumping units taking into accounts technical characteristics of the power plants facilities (turbine/generator and pump/motor systems), reservoirs characteristics considering all hydraulically connected infrastructures which are, as well as the hydrological conditions (for run offs and turbine inflows). The goal of the operation for all hydropower system is to have maximum profit from electricity production taking into account electricity market environment, power system requirements, as well as satisfying technical conditions and at the same time avoiding water spilling as it is maximum possible. Operating modes of the hydro power plants depend on electricity market price, tariff prices and time duration of each tariff (base, peak, and rush hour), technical characteristics of the unit(s), hydro technical conditions, water reservoir’s volumes, run off and others. The application of the model is presented with the results on real hydro power systems consist of different hydraulic connections in various cases depend on hydrological conditions and requirements. The model can be used to make preliminary analysis for additional improvement of hydro systems as having the results for additional installed power of the existing system. The selected output results are presented with appropriated comparing and at the end with conclusion remarks.