A model of a Micro Gas Turbine system for cogeneration is presented. The analyzed plant is based on an aero derivative Gas Turbine with a single staged centrifugal Compressor and an axial Turbine with two stages. The net power output is 260 kWe in simple cycle mode. Exhaust gases can be sent to a counter flow surface compact heat exchanger for thermal regeneration, which turns to be thermodynamically favourable in this range of power output. If a thermal load is required the system operates in CHP configuration and part, or the whole, of turbine exhaust gases are sent to a Heat Recovery Boiler for water heating. The HRB is, in analogy to the Regenerator, a counter flow surface heat exchanger. The mass of hot gases directed to each heat exchanger can be controlled by a regulation valve that allows, for a given fuel mass flow rate, to enhance the net power output or to privilege the thermal generation at the HRB. This degree of freedom allows the system to operate at different cogeneration degrees, thus covering many power-to-heat demand ratios. The whole system is modeled in the Simulink® environment, a powerful tool for dynamic system analysis. All components are studied and a mathematical representation for each of them is described. Equations are then implemented in Simulink® allowing to create customized blocks of different components which are then properly coupled, respecting the physical causality of the real system, by connections that may represent either mechanical or fluid dynamic links. Models are classified depending on whether state variables for the considered component can be defined or not. Compressor and turbine are represented as “Black Box” components without state, while the combustion chamber is modelled as a “white box” applying energy and mass conservation equations with three state variables. Heat exchangers are considered as “White Box” without state, and the physics of the heat exchange process is studied according to the Effectiveness-NTU method. A further dynamic equation is the shaft dynamic balance equation. Model results are reported in the paper in several transient conditions: in all cases the computational time proved to be lower than real time.

This content is only available via PDF.
You do not currently have access to this content.