Optimal unit commitment with fuel, storage, and environmental constraints

The purpose of this research effort is the general formulation and the development of a dynamic programming based algorithm to optimally solve the short-term unit commitment problem in electric power systems with fuel and environmental constraints.;The objective is the minimization of the cost of operation over a prescribed time horizon. Two types of fuel constraints are considered: those imposed by fuel storage and those resulting from the fuel delivery system. The emission constraints are those imposed by legislated limits on air pollution. Also included are all the standard constraints, such as unit capacity limits, minimum system spinning reserve, unit minimum-up and minimum-down time, crew constraints, etc.;Another important goal for this study is to produce an algorithm that will render optimal generation schedules efficiently, both in terms of computer time and memory, to permit their successful implementation.;The fuel constrained unit commitment problem is formulated as a three resource problem. The first involves the storage of fuel with prescribed maximum and minimum storage levels, under a constant supply of fuel. The dynamic programming formulation requires storage parameterization for this type of resource. The second includes a fuel pipeline delivery system with contractual obligations. These are translated into maximum and minimum hourly fuel delivery constraints, and maximum and minimum daily delivery constraints. The third resource is assumed to be plentiful and thus unconstrained.;The complexity of the problem is greatly simplified by the transformation of the fuel delivery constraints into unit capacity limits using a closed-form dual dispatch approach. The virtue of this transformation is that it renders the fuel delivery constrained units as fuel unconstrained. The dual-dispatch strategy calls for the storage constrained units to be dispatched to maximize the power they generate for a binding amount of fuel or fuel cost, whereas the remaining unconstrained units are economically dispatched to provide for the remaining power needed to meet the demand.;The emission constrained problem is a generalized formulation that not only addresses the implementation of the two-phase requirements of the U.S Clean Air Act Amendments of 1990 (CAAA), including its provision for emission allowances trading, but also allows for any additional locally legislated requirements. Thus the constraints are prescribed on multiple pollutants as are typically found in the combustion of fuel. For each pollutant the constraints include the maximum hourly emission on every unit, the maximum hourly emission on every set of on-line units, and the maximum daily emission for the system.;For both the fuel and the emission problems, necessary and sufficient conditions are established and effectively used for the a priori (or off-line, that is before the start of the unit commitment algorithm) elimination of all the infeasible unit combinations that cannot meet the system constraints, thus greatly increasing the algorithm efficiency.;Test results are provided to illustrate the merits of the proposed methods.