Short-Term Generation Schedule Optimisation for Combined Heat and Power
In the last decades one of the most difficult problems in the electricity market has been how to dispatch and manage the electricity in power generation plants. Despite of all the benefits of distributed poly-generation and combined heat and power systems, their penetration in the power market worldwide is quite modest and one of the barriers against their increasing participation is the high fees for back-up supplies, which is one of the problems addressed in this investigation. This paper introduces a pool of distributed generation units (named nerve-centre) able to economically optimise the generation schedule of gas turbine power plants and end-users interconnected through a mini-grid. A hybrid genetic algorithm adapted priority list was developed to solve the multi unit generation schedule optimisation problem. The algorithm developed in this study leads the optimisation mechanism to a faster convergence and a very low risk of non-convergence to the optimal result. Despite the power generation optimisation studies reported in the technical literature, none of them has been modelled for such a pool of distributed generators trading electricity in the competitive market. This investigation shows that the proposed nerve-centre concept can result in significant savings to generators/end-users.