Compressor Instability Analysis Within an Hybrid System Subject to Cycle Uncertainties

The transient/dynamic modeling of energy systems can be used for different purposes. Important information can be obtained and used for the design phase. The control system and strategies can be safely tested on transient/dynamic models in simulation, increasing the confidence during experimental phase. Furthermore, these models can be used to acquire useful information for safety case. The analysis of energy systems in dynamic conditions is generally performed considering fixed values for both geometrical and operational parameters such as volumes, orifices, but also initial temperatures, pressure, etc. However, such characteristics are often subject to uncertainty, either because they are not known accurately or because they may depend on the operating conditions at the beginning of the relevant transient. With focus on a micro-gas turbine fuel cell hybrid system, compressor surge may or may not occur during transients, depending on the aforementioned cycle characteristics; hence compressor surge events are affected by uncertainty. In this paper, a stochastic analysis was performed taking into account an emergency shut-down in a fuel cell gas turbine hybrid system, modelled with TRANSEO, a deterministic tool for the transient and dynamic simulations of energy systems. The aim of the paper is to identify the main parameters that impact on compressor surge margin. The stochastic analysis was performed through the RSA (Response Sensitivity Analysis) method, a sensitivity-based approximation approach that overcomes the computational burden of sampling methods such as MCS (Monte-Carlo Simulation). The results show that the minimum surge margin occurs in two different ranges of rotational speed: a high-speed range and a low-speed range, the latter being more sensitive for surge occurrence. The temperature and geometrical characteristic of the outer pressure vessel, where the fuel cell is installed, are the two main parameters that affect the surge margin during an emergency shut down.