The influence of the instantaneous fuel mix for electricity generation on the corresponding emissions

Abstract

Since it would be unwise to apply an environmental measure before its impact is estimated, tools are needed to simulate the system to which the measure is applied. In this paper, a tool and a methodology are discussed in order to simulate scenarios for demand and supply side options and to quantify the emissions and the energy use related to electricity generation. To supply the electricity demand, the composition of the active power system evidently changes with the variations in demand. The instantaneous composition of the power system is determined by the activation order (set by a chosen criterion) of all power plants. Because of their specific nature, some problems are dealt with separately in the model. Pumping plants are both customer (pump-mode) and provider (turbine-mode) of electric power. Therefore, they are not modeled as other plants, but as an adaptation in the demand pattern for other plants. Outages are mostly unpredictable but very important for the instantaneous energy use and emissions. Therefore, the model contains a component in which an average outage schedule can be calculated. The possible deviation of this average result is dealt with in the calculation of the uncertainties. Also, other parameters, like accidental unavailability of plants, human decisions in activation and deactivation of plants and the unpredictable availability of cogeneration units, are included in the calculation of the uncertainties. A simulation tool is necessary, because a-priori “linear” estimates do not always appear to be accurate. Results are highly dependent on the composition of the power system and the activation order of the individual plants. Demand and supply side actions are therefore simulated. Simulations for Belgium have pointed out that variations in electricity-related greenhouse-gas emissions (per unit of electricity generated) can be larger than a factor two, depending on the time segment in which the electricity is consumed. For SO x and NO x emissions this variation can be even larger.