Toward the concept of integrated security: optimal dispatch under static and dynamic security constraints

Abstract

No analytical method is yet known which can solve the optimal power flow problem, off-line or on-line, when it is constrained by both static and dynamic security. The concepts introduced in this paper can, however, contribute substantially toward obtaining on-line solutions of the problem. Key ideas behind the concept of an integrated security constrained optimal dispatch (ISCOD) are presented. Its purpose is to calculate the power outputs of generators and the voltage magnitudes of buses so that the cost of power generation is minimized and both static and dynamic security of the system are maintained. While analytical tools for steady-state security constrained dispatch are available, those for dynamic security constrained optimization are not well defined. In addition to this, on-line implementation of either function presents an entirely new gamut of problems. ISCOD combines the diagnostic and descision-making capabilities of a knowledge based system (KBS) and the massive parallelisms and learning features of an artificial neural network (ANN) along with conventional power network solution methodologies to provide real-time optimization and control. The real-time functions that are required in ISCOD are: (i) contingency selection and ranking for steady-state security, (ii) fast static security assessment, (iii) contingency selection and ranking for transient stability and subsequent dynamic stability assessment, and (iv) control measures for preventing security violations.