Page 145 - Advanced Thermodynamics for Engineers, Second Edition
P. 145

132    CHAPTER 6 FINITE TIME (OR ENDOREVERSIBLE) THERMODYNAMICS




                                1.0
                               Temperature ratio of low temperature engine  0.8

                                0.9





                                0.7


                                0.6

                                0.5
                                  0.5      0.6     0.7      0.8     0.9      1.0
                                          Temperature ratio of high temperature engine
             FIGURE 6.8
             Variations in temperature ratio of high- and low-temperature heat engines that produce maximum power
             output.

                Thus the variation of s 1 with s 2 to produce maximum power output is shown in Fig. 6.8; all of these
             combinations result in the product being 0.5.


             6.4 PRACTICAL SITUATIONS

             The ideas put forward in this chapter may seem to be rather esoteric, and without much connection to
             the real world. However, it is well known that ‘real’ engines never achieve the Carnot efficiency, and
             Bejan (1988) compared data from power stations to that which would be predicted using Eqn (6.14);
             this is shown in Fig. 6.9. It can be seen that the ‘best’ power stations lie on the line depicted by Eqn
             (6.14). Most of the data lie below that line, and this could be because the actual ‘heat engine’ (i.e. the
             plant excluding the heat transfers into and out of the system) is nowhere near reversible, or the
             operating condition is not at the maximum theoretical power. Either way, this approach goes some way
             to showing why there is such a big difference between the Carnot efficiency and that obtained in
             practice, and enables a more realistic assessment of the potential efficiency of a power plant.
                It is also possible that some engines will achieve efficiencies that lie above the line, and these might
             be thought to show the analysis is incorrect. Gyftopoulos (1999), in a criticism of FTT, gives the
             example of a combined cycle power plant achieving an overall thermal efficiency of 60%, which he
             states would imply a temperature ratio, T H /T C ¼ 6.0, and results in T H ¼ 1800 K – obviously an un-
             reasonable value. However, by reference to Fig. 6.2, such an efficiency of 60% is achievable if the
             criterion of maximum power is relaxed. Calculations with T H /T C ¼ 4.0 show that the loss of power
             operating at T 1 /T 2 ¼ 2.5, to achieve the 60% efficiency, rather than T 1 /T 2 ¼ 2.0, results in a 10% loss of
             power, but a 20% gain in efficiency. Hence, sometimes operators will choose to operate at a higher
             efficiency than achievable at maximum power; this particularly true when fuel prices are high.
   140   141   142   143   144   145   146   147   148   149   150