Steady-state turbogas plant performance optimisation through mean-line and stream-line analyses of turbine and compressor

This thesis project takes place in the context of heavy-duty gas turbines production, evolution and upgrades. Given the trends going more and more towards higher output powers in order to satisfy the energy demand, increased efficiencies and lower emissions to deal with the stringent regulations imposed by the Countries concerning environment on the one hand and the necessity to guarantee a machine life as long as possible to amortise its capital cost on the other one, the partial re-design of old generation machines to make them competitive on the market with new generation ones becomes predominant. The focus of the present thesis is centred on a machine, the FIAT TG20 B7/8, whose original design dates back to the Seventies, moreover many turbo-gas plants all over the world still present this machine. The main objective is to carry out an analysis of the machine able to pave the way for all the modifications necessary to increase power and efficiency, identifying the most important parameters to act on, perform simple and low-cost yet effective and well-studied changes and deal with the possible critical issues which could be generated from the latter. The starting point has been a series of sensitivity analyses, conducted from a mean-line standpoint, in order to understand the behaviour of the machine, the final assessment has been to increase the turbine inlet temperature by an amount consistent to the turbine components materials resistance and to reduce the cooling mass flow rate. Even though this action may appear as a countersense, it must be specified that in the original design a not so high attention was paid to all the possible leakages which could come from the cooling circuit, therefore the idea behind that has been to reduce the cooing mass flow rates where not needed, to reduce the leakages and to keep the necessary ones possibly unchanged coherently to the mechanical part creep and overtemperature resistance. Consequently the cooling system has been deeply analysed in order to find the critical spots and strategies able to reach the above-mentioned target, still applying slight modifications on the mechanical parts, which do not represent a problem from the manufacturing, mounting or cost point of view. The last part consists instead in a sort of validation of all the work previously done, in fact the same machine has been studied by using two completely different soft-wares. One of them, which dates back to the Seventies as well, is able to perform the analysis just from a mean-line standpoint and all the characteristics of the machine have to be input manually, starting from drawings. The other one, a more modern software, is able to perform both a mean-line and a streamline analysis where the geometry of the machine can be directly acquired from 3-D CAD models. The stream-line approach becomes then important for the thermo-structural validation of the work. The two of them have been compared from a mean-line point of view, leading to outputs which deviate by a small, fixed amount, considering different turbine inlet temperatures and cooling conditions. All the considerations and analyses made in this work will then lead to a relatively new machine, able to compete with new generations ones.