The environmental protection policies and legal obligations motivate process industries to implement new low-emission and high-efficiency technologies. For the purpose of production process optimization and related control system design it is worthwhile to first build an appropriate process model. Apart from favorable execution speed, accuracy, and reliability features, the model also needs to be straightforward and only include the physical and design characteristics of the overall plant and its individual components, instead of relying on empirical relationships. To this end, this paper presents a nonlinear dynamic model of the single-drum natural-circulation steam boiler evaporator circuit, based exclusively on the fundamental physical laws of conservation of mass, energy and momentum, wherein the reliance upon empirical relationships has been entirely avoided. The presented boiler system modeling approach is based on the analysis of the physical phenomena within the boiler drum, as well as within downcomer and furnace tubes, and it also takes into account the boiler system design-specific features such as cyclone steam separators, thus facilitating the derivation of a fully-physical process model. Due to the straightforwardness of the derived process model, it should also be useful for the analysis of similar steam boiler facilities, requiring only adjustments of key operational and design parameters such as operating pressure, temperature, steam capacity and characteristics of ancillary equipment such as pumps. To illustrate the model effectiveness, it has been employed in the analysis of the phenomena occurring in different parts of the particular boiler system for the case of realistic disturbance event, wherein the model inputs are based on the field data from the boiler on-board data collection system (DCS). It is anticipated that the proposed physical boiler model should also be easily adapted for the case of boiler systems utilizing alternative fuels, thus aiding in the optimization of the dedicated control and supervision systems. (c) 2014 Elsevier Ltd. All rights reserved.