Security Regions Based Hydrothermal Optimal Power Flows

The Optimal Power Flow (OPF) is used extensively in the control centers of power utilities as well as by system planners for planning studies, as it calculates various levels of power generation and all the controls associated with the static (steady-state) operation of the entire electric power grid of a power utility. During the course of development of the OPF, it has been realized that the main challenges relate to the effectiveness of dealing nonlinear functional constraints, the modeling of the problem, and the speed and reliability for simultaneously solving all of the network equations while minimizing an objective function. The largest challenge in OPF is how to deal with transient stability and static voltage stability constraints at the same time.A security region (SR) defined in injection space is unique for a given power system configuration or a power system with a particular postulated change in configuration. Different from traditional"point-wise"approach, a security region defined mathematically and visualized can give power system engineers systematic and global information about the feasible operation region. Two type of security have reached the stage of practical application: one is the practical dynamic security region (PDSR) in power injection space to guarantee transient stability, the other is the static voltage stability region (CVSR) in critical cut set power space. Based on the hyper-plane descriptions of critical boundaries of PDSR and CVSR, the difficulties in dealing with stability constraints in the problems of solving OPF may be overcome easily.Firstly in this dissertation a novel mathematics model and an efficient algorithm that can minimize total generation cost with considering both 1st order and 2nd order terms of generation cost function, are developed, where both of transient stability constraints and static voltage stability constraints are taken into account based on the hyper-plane descriptions of critical boundaries of security regions. In the formulation of OPF an affine mapping between branches angles and real powers is used to take branch angles as optimization variables instead of power injections as usual. For simplicity, an affine mapping between voltage magnitudes of buses and reactive power injections in network is also used. The feasibility and effectiveness of the proposed mathematics model and algorithm have been validated in the 39-bus New-England System.The formulation of the optimal hydrothermal power flow is one of fruitful applications of OPF, which is more realistic than conventional OPF because of dynamic coupling between the variables of the problem as a result of the hydro energy constraints introduced through the volume of water availability limitations.Secondly, in this dissertation a novel formulation of the optimal hydrothermal power flow problem is also suggested with taking into account the emissions minimization, and the transient stability and voltage stability in its constraints based on the concepts of security regions. The transmission loss is approximately expressed in terms of the generalized generation shift distribution factor and of generated power. The implementation is based on a Newton Raphson's interactive procedure, with novel initial guesses to obtain improved convergence properties. The IEEE standard systems are worked out in order to demonstrate the effectiveness of the proposed method.Finally, the security region constrained distributed optimal power flow for interconnected power system is presented in this dissertation. The centralized OPF problem of the multi-area power systems is decomposed into independent distributed OPF subproblems one for each area. The transient stability constraints and static voltage stability constraints, and line current limits are included as constraints. The solutions of the distributed OPF subproblems make the independent dispatch of each area possible while achieving a multi-area optimum. The nonlinear distributed OPF subproblem is solved by an efficient predictor-corrector interior point method. The IEEE three-area RTS-96 system is worked out in order to demonstrate the effectiveness of the proposed method.