Fiche individuelle

This paper presents a general operational planning framework for controllable generators, one day ahead, under uncertain re-newable energy generation. The effect of photovoltaic (PV) power generation uncertainty on operating decisions is examined by incorporating expected possible uncertainties into a two-stage unit commitment optimization. The planning objective consists in minimizing operating costs and/or equivalent carbon dioxide (CO 2 ) emissions. Based on distributions of forecasting errors of the net demand, a LOLP-based risk assessment method is proposed to determine an appropriate amount of operating reserve (OR) for each time step of the next day. Then, in a first stage, a deterministic optimization within a mixed-integer linear programming (MILP) method generates the unit commitment of controllable generators with the day-ahead PV and load demand prediction and the prescribed OR requirement. In a second stage, possible future forecasting uncertainties are considered. Hence, a stochastic operational planning is optimized in order to commit enough flexible generators to handle unexpected deviations from predic-tions. The proposed methodology is implemented for a local energy community. Results regarding the available operating reserve, operating costs and CO2 emissions are established and compared. About 15% of economic operating costs and environmental costs are saved, compared to a deterministic generation planning while ensuring the targeted security level.

Abstract In electrical systems, the main objective is to satisfy the load demand at the least cost without having imbalance between generation and consumption. Thus, the uncertainty of photovoltaic (PV) power production must be considered in generation planning. In this paper, the optimal generation scheduling in an urban microgrid is made by taking in consideration operating reserve (OR) provision and under stochastic characteristics of PV power prediction. By considering a prescribed risk level of unbalancing, a dynamic programming (DP) algorithm sets the operational planning of conventional generators, so that the operational cost and available operating reserve can be calculated. Then, the effect of PV power uncertainty into the unit commitment is analysed by considering PV forecast intervals with a 95 % confidence level. The unit commitment is then recalculated with new generator set points and the same criteria. Finally, variations of the targeted minimized costs and obtained OR is analysed according to the considered uncertainty.

High levels of renewable generation incorporation increase the flexibility requirements of the electrical system in response to fast and large variations in load and renewable energy output. Advanced power system operational methods and algorithm are required to schedule the generating units with more reliability and efficiency. In order to deal with uncertainties in generation planning of an urban microgrid, this paper presents a storage control strategy for reserve provision in a multi-objective scenario-based stochastic optimization algorithm. Power reserve allocation is optimized while uncertainties from renewable energy and load demand are taken into account. Results confirm that the presented algorithm ensures that storage-hybridized RESs contributes to the primary source of reserve capacity, and the performance in terms of security level are highlighted.

In electrical systems, the main objective is to ensure that users demand is met at the least cost without having imbalance between generation and consumption. Thus, the uncertainty of photovoltaic (PV) power production must be considered in generation planning. In this paper, the optimal generation scheduling including the operating reserve (OR) provision are developed under stochastic characteristics of PV renewable energy in an urban microgrid. With a prescribed risk level of unbalancing, a dynamic programming (DP) algorithm sets the operational planning of conventional generators. Then, the operational cost and available operating reserve can be calculated. The effect of PV power uncertainty into UC is then analyzed by considering forecast intervals of PV forecasting. The proposed methods consider PV prediction uncertainties with a 95% confidence level. The unit commitment is then recalculated as well as new generator set points with same criteria. Hence, variations of the targeted minimized costs and obtained OR is analysed according to the considered uncertainty.

Electrical system operators and automatic controllers use Operating Reserve (OR) power to mitigate the unexpected imbalance between power supply and load demand. This backup power should be carefully sized and dispatched to reduce operation costs while keeping a satisfying security level. With the integration of small sized intermittent renewable generators in distribution networks, the allocation of OR where they are connected is interesting in order to compensate directly these uncertainty sources and maintain the security and reliability. In this paper, we consider the provision of OR directly by distributed PV generators combined with energy storage systems. For an urban microgrid, comparisons are given with the OR provision by a micro gas turbine. A method for dynamic joint dispatching of OR power on both generator types is presented and tested. Results show new insights in the possibilities of OR dispatching with renewable energy sources.

In electrical systems, the unit commitment (UC) and power scheduling plan the operating of generating units in order to satisfy the load demand under system operating constraints. Nowadays, energy communities have emerged with individual community energy requirements and increasing capacity deployment of distributed energy resources. The high penetration of renewable energy sources (RES) and load demand increase locally the power system uncertainty. Hence, traditional deterministic approaches for one day ahead UC should evolve to stochastic optimization methods. The main goal of this thesis is to propose a probability-based and stochastic optimization methodology for optimal generation and operating power reserve (OR) scheduling decisions in an urban microgrid, with the objective of addressing the minimization of operating costs and emissions. Based on an uncertainty modelling with forecasting error distributions, a LOLP-based risk assessment method is used to determine an appropriate amount of OR for each time step of the next day. Then, in the first stage, a deterministic optimization within a mixed-integer linear programming method generates the unit commitment of controllable generators with the day-ahead PV and load demand prediction. In the second stage, a set of scenario is built to model future and probable uncertainties. It is integrated into a stochastic optimization of the operational planning. Issues of the second stage are the commitment of enough flexible and fast generators to handle unexpected deviations from predictions. In order to decrease emissions, the scheduling and operational planning of local storage systems for OR provision are considered ; the PV self-consumption is increased and operational cost are decreased. The significance of the proposed methodology is illustrated with results obtained from a studied urban microgrid system. A user-friendly Supervisory Control And Data Acquisition system is developed with the Matlab GUI to integrate and visualize the energy management operation. Mots-clés : Stochastic optimization, uncertainty, unit commitment (generation scheduling), energy management system, power reserve, renewable energy, microgrid