Dynamic economic dispatch with sensitivity analysis

Abstract

The thesis considers the sensitivity analysis of the Dynamic Economic Dispatch Problem. This involves the variation of the load demand and the cost coefficients of the objective function. The objective function to be minimized is considered to be the sum of the production cost functions of the thermal units (already committed) in all time intervals. The Dynamic Economic Despatch Problem is solved using a redispatch technique. The latter is formulated considering the load demand, unit limits, system spinning reserve requirement, security constraints of some of the transmission lines and power rate limits of the units. The redispatch technique linearizes the unit outputs about the base case static economic dispatch (SED). The SED solution is obtained using a quadratic programming algorithm in this thesis, although any conventional SED algorithm will be suitable. After that the dispatch problem is linearized about the optimum points obtained from the SED solution, a linear program (LP) is used to calculate the changes in the units output. The LP solution is combined with a parametric program (PP) in order to be able to carry out sensitivity analysis for the LP solution. In this case the LP solution provides dynamic information. A static optimum solution will become obsolete as soon as parameters of the LP model are changed.