A suitable probabilistic scenario set of load demand and natural characteristics of renewable energy is becoming a crucial issue in power system planning studies. Properly addressing the impact of potentially thousands of residential PV plants on the resilience and reliability needs of substations necessitates the representation of inherent relations between photovoltaics and the load throughout the long-term planning period. The optimal planning of substation expansions is achievable through proper modeling of input parameters which describes the characteristics of the service areas. In this paper, the co-existence of PV plants and the load in a service area under three different states such as daytime with clear-sky and no-fault, daytime with abnormal events, and nighttime are incorporated into the stochastic dynamic optimization problem by using scenario-based approach. The scenario tree of the problem is branched from three different bases simultaneously instead of only one as in conventional approach. This paper also combines the risk-constrained stochastic dynamic SEP problem and Mixed Integer Linear Programming (MILP) framework under one roof. The comparison between integrating inherent characteristics of PV plants with and without considering abnormal events into the optimization is performed to show the impact of suitable probabilistic model on dynamic nature of investment decisions.

Research on monitoring and fault detection systems for photovoltaic plants is significantly increasing with the continual development in technologies and the availability of qualified data. Nevertheless, many gaps still exist that need to be addressed. The electrical output of PV plants in various weather conditions is very close to those obtained under fault conditions. For a large scale PV plant, it is very crucial to utilize the data in the decision-making process without using external sensors or performing simulation studies that require detailed parameters of the plant. The main challenge here is to automatically rationalize the collected data in order to make a decision on distinguishing between faulty and natural outputs. This paper proposes a method for distinguishing faults and inherent changes in the PV plant’s output to help O&M crews identify and fix system issues. The proposed method has the ability to map the inherent characteristics of the PV plant by using only the data received from inverters without using additional equipment or detailed models. It has been developed by analyzing the working mechanisms of several large scale PV plants installed in Turkey. The proposed method can easily be implemented in a newly installed or existing PV plant. The novelty of this study is detecting abnormal operations in a PV plant even under low irradiance and cloudy-sky conditions without using any irradiance and temperature sensors. The effectiveness of the proposed method is shown in rooftop and ground-mounted PV plants. 

Renewable energy sources (RES) are increasingly integrated into power systems, introducing operational uncertainties that challenge the way we manage the grid. These uncertainties necessitate strategies to address the risks in grid reliability and economic performance. This paper introduces a framework to simultaneously manage the risk associated with economic performance and grid reliability under non-Gaussian uncertainty. The framework utilizes Polynomial Chaos Expansion to solve the risk-based and chance-constrained Stochastic Optimal Power Flow for hybrid AC/DC grids. The risk associated with the costs is addressed by introducing the Value-at-Risk parameter, derived through moment-based calculations, to facilitate risk-averse decision-making. Numerical studies illustrate the impact of risk-neutral versus risk-averse decision-making on the probability distribution functions of RES curtailment and operational costs. Additionally, analyzing efficient frontiers for various confidence levels showcases the framework's capability to construct a portfolio of strategies that effectively balance risk and operational costs under varying confidence levels of non-Gaussian uncertainty.

In this paper we are presenting a preventive security constrained optimal power flow model for AC/DC grids using a robust optimisation of converter active power-voltage droop coefficients and generators and HVDC converters active and reactive power set points. The converters’ optimal droop control actions maintain system feasibility after DC grid contingencies. The model takes into account wind power and load demand uncertainties and uses a scenario based approach to robustly determine the HVDC converters’ voltage-power droop coefficients. The developed optimisation model is applied to a variety of AC/DC grid test cases for the analysis of its computational performance, convergence and AC feasibility of the obtained solution. Finally, we propose a number of future extensions and modelling improvements for real-life applicability of the developed model in the day-ahead operational time frame.

Unprecedented events may cause power system failures and blackouts whose impact may be highly destructive. It is therefore of extreme importance to design and operate power systems that ensure high resilience and efficient restoration plans in case of blackouts. While renewable energy sources (RES) are invaluable assets for a sustainable future, the uncertainty present in RES-dominated systems poses a challenge in developing optimal restoration strategies. We address this challenge in determining optimal generator start-up sequences through a novel mathematical formulation and solution methodology that incorporate the uncertainty in renewable participation by means of chance constraints. The novel formulation outperforms the state-of-the-art formulation for solving the deterministic problem for medium- and large-scale instances as well as provides the optimal restoration time in a matter of seconds on a medium-scale network for the stochastic problem. We further provide insightful observations with respect to renewable contribution under different confidence levels and optimization objectives.

Utilizing reliable monitoring and available data in order to ensure that the PV plants become best-return-on investments is possible thanks to ever going developments in data-driven algorithms. One of the major challenges in PV power generation is the soiling events caused by the dusty environments and climatic conditions resulting in the reduction of the performance of the PV plants. Since the soiling is a complex event whose characteristics are forged by a number of external stochastic phenomena, a number of studies are conducted by researchers. Soiling in PV plants is a continuous subject that emerges over time as an important drawback in the performance of the PV plants. In that point, it is crucial to automatically determine appropriate cleaning schedules in order to achieve a high-profit PV plant operation. In this study, a method is presented which only uses the DC current inputs of the inverters to determine the soiling loss without using any environmental sensors. The proposed method is tested by conducting experimental works on three different PV plants which are connected to the power grid in Turkey. The results show that the proposed method is effective for analyzing the performance reductions due to soiling in the PV plants without using any environmental sensors.

 Making decisions on expansions of the power system as a result of the growth in electricity demand is essential in order to maintain a reliable network. Since substations take on the role of the intermediate link between electricity supply and demand, substation expansion planning (SEP) is one of the most important planning studies. In this study, the SEP problem is merged with the different geographical configurations of the zones within a mixed integer linear programming framework. The voltage profile and power losses are integrated into the SEP problem by using approximate lumped model of the zones. The effect of voltage constraints and the cost of feeder loss on the SEP problem is investigated and the comparison between the SEP problem with rectangular zones and with triangular zones is performed to show the impact of different geographical configurations on the investment decisions.

The operational management of the Belgian transmission system is increasingly challenging due to the increased penetration of renewable energy sources and the European electricity market integration. DISCRETE will answer fundamental research questions on the applicability of new data driven optimization models and uncertainty modelling to minimize the total cost of operation and CO2 emissions. DISCRETE will allow the development of new decision support tools, enabling secure power system operation. (Link) 

Researcher at TUBITAK (Scientific and Technological Research Council of Turkey) 1507 Project, Grant No 7190391, Project Title: "Solar Power Plants Remote Monitoring, Control and Fault Diagnosis System".

• Developed intelligent methods for fault detection and diagnosis in solar energy systems. 

A suitable probabilistic scenario set of load demand and natural characteristics of renewable energy is becoming a crucial issue in power system planning studies. The optimal planning of substation expansion is achievable through proper modeling of input parameters which describes the characteristics of the service areas. In that point, the disparity between the required resources for adequate and flexible power systems and the budget available to planners necessitates choosing between different expansion strategies. Under these circumstances, making investment plans with a convenient risk management strategy is crucial in order to maintain the balance between the cost and system adequacy. In this thesis, the SEP problem is handled from a risk management point of view by integrating different investment strategies. On the one hand, the co-existence of PV plants and the load in a service area under three different states such as daytime with clear-sky and no-fault, daytime with abnormal events, and nighttime are incorporated into the stochastic dynamic sub-transmission SEP, in which the decision-maker chooses the risk-neutral and risk-averse strategies, through systematic probabilistic scenario tree generation. The comparison between integrating inherent characteristics of PV plants with and without considering abnormal events into the optimization is performed to show the impact of a suitable probabilistic model on the dynamic nature of investment decisions. On the other hand, the distribution SEP problem in which the decision-maker chooses the risk-seeker strategy is merged with the different geographical configurations of the zones within a mixed-integer linear programming (MILP) framework by using approximate lumped models. 

The aim of probabilistic power flow is to identify the possible ranges of power flow, bus voltages, and bus angles. In that manner, the inputs and system variables of the power system are considered as random variables with particular probability distributions. In this project, different approaches for probabilistic power flow will be implemented and compared in the MATLAB environment considering wind and load powers through IEEE test networks. The application of the probabilistic power flow techniques in power system studies is also discussed.