Optimization of the Operation of Isolated Industrial Diesel Stations

The purpose of this work is to minimize off-grid Diesel generation costs by best choosing the number, size and dispatch schedule of the generators comprising the power plant. The mathematical formulation of the optimization problem takes into account: sizedependent full-and part-load fuel consumption; size dependent investment costs; sizeand dispatch-dependent operational and maintenance costs. It is assumed that daily hourly demand is known. The objective function is the annual sum of the fuel, investment and operational and maintenance costs on an hourly basis. The decision variables are the hourly generator’s load factors. In the optimization routine, the number of generators is prescribed. Generators are chosen within a prescribed power range from a data bank of industrial diesel gensets. All possible combinations of the generators are optimized. Possible combinations are the ones that meet the hourly demand and may include duplicated generators. The developed program lists the best five combinations. To solve the optimization problem we have used the available NCONF/IMSL subroutine, which is based on the successive quadratic programming method to solve general nonlinear optimization problems. The optimization problem was applied to three north-region Brazilian communities. For these sites, we have concluded that substantial reductions on generation costs can be accomplished through generators resizing and through appropriate maintenance.