4 Comparison of various gas turbine cycle architectures

Page 1

Fourth project: comparison of various gas turbine cycle architectures Simone Fedeli 892133 Luca Grippo 829683 Filippo Mecarelli 892120 Francesco Persico 893256 November 2017 1 The purpose of this analysis is to compare the performance of four dierent gas turbine architectures: • simple cycle; • intercooled cycle (with βCLP = 3 √ βCtotal ); • recuperative cycle; • intercooled-recuperative cycle (with βCLP = √ βCtotal ). 1 General performance analysis of the four cycles TIT=1200 ° C, xed. 1.1 Simple cycle From the theoretical analysis of the simple Joule-Brayton cycle, which does not include technological peculiarities (most importantly cooled expanders), we know that: • η = 1−β 1−γ γ = 1−β−θ; • l =c0 P (TIT (1−β−θ)−T0(βθ− 1)); in words, this means that theoretically e ciency always grows with β, reaching Carnot's e ciency when the compression ratio reaches βmax, i.e. the theoretical compressor outlet temperature is the same of the turbine inlet temperature; intuitively, when these two temperatures are the same, the fuel mass ow rate is null, so the work/power of the cycle is null too. So, while e ciency is monotone with β (always growing), the speci c work has a maximum (which theoretically is at βmaxwork =√βmax), and then comes back down to zero. The theoretical 2 performance line in the plot above should be, then, a sort of horizontal parabula (concavity to the left). Using a software that simulates much better the real thing, we can see that the real performance is very dierent from the one predicted by the theory at medium and high compression ratios: the real line becomes at (in terms of e ciency), even decreasing a little bit after β = 32, while also the speci c work decreases very sharply starting with β = 28 ; the theoretical model does not account for…