Desenvolvimento e avaliação de metodologias para máxima alocação de geração eólica em SEP e controle de frequência no fluxo de potência CC

Abstract

This work is divided into two studies applied to Electric Power Systems (EPS) in steadystate assessment. The first one is an analysis of the impacts of wind power allocation on EPS and the second one is a study of the application of primary frequency regulation in the linearized power flow problem. In the first study, three formulations are proposed to evaluate the choice of the bus for the installation and sizing of the installed capacity of wind generators in EPS. Equations describing the intermittent wind generation behavior and the load variation in the three formulations are considered. The first proposed methodology is based on fully linear programming (PL–CC) approach using a linearized power flow model, the second is a non-linear optimization model considering a non-linear power flow (PNL–CA) formulation, and the third one is an integer nonlinear programming approach (IPNL–CA). The three methodologies developed are able to safely estimate the maximum penetration level of wind power in the grid. The wind power penetration level is the maximum installed power (MW) of the wind turbine(s) since it is not considered to store this energy by the three methodologies, which respect the load/generation balance and the system limits of security. The methodology PL–CC and PNL–CA aim reaches this maximum level by (MW) wind turbines pre-installed in the system. The formulation IPNL–CA performs a sizing analysis in conjunction with the allocation, choosing the ideal bus to be installed in wind generation, to reach the maximum level of safe penetration. The results obtained show the advantages and disadvantages of the use of each formulation proposed in the analysis of wind power generation in electric power systems. In the second study, the steady-state characteristics of the primary frequency regulation of generators are combined with a new formulation of the DC power flow model. The proposed approach consists of including the equations representing the steady-state response of the primary frequency regulation in the original DC power flow modeling. Thus, the imbalances between load and generation are distributed among the generators as a function of the speed-droop and the reference power of each machine. The proposed methodology is validated in three systems: two are small-scale and one medium-scale test system. The first one is a 6-bus tutorial system, the second is the 11-bus system with two distinct areas, and the third is the 39-bus New-England power system. The results presented validate and indicate the effectiveness of the proposed formulation.