Vol.5 No.4 2013
31/70

Research paper : Evaluating Uncertainty for the Standardization of Single Cell/Stack Power Generation Performance Tests for SOFC (A. Momma et al.)−255−Synthesiology - English edition Vol.5 No.4 (2013) input quantities, thereby minimizing the need to take into account any possible correlation between input quantities for uncertainty analysis. The draft standard additionally defined the maximum allowable uncertainties of measurement instruments (i.e. uncertainty due to systematic error) (Table 1). This was specifically intended for the uncertainty of the instruments at the time of calibration. However, the draft standard also allowed a simplified Type B evaluation that assumes a uniform distribution, using catalog values such as the degree of accuracy of the instruments, on the condition that the instruments have been calibrated.4 Uncertainty of measurement of SOFC properties4.1 Selection of factors affecting uncertainty through experimental verificationThe properties of SOFC depend on various factors such as cell temperature, fuel flow rate, air flow rate, operating pressure, and temperature distribution. It follows that the uncertainty of the control and measurement of these input quantities would inevitably have an effect on the uncertainty of the measurand. We decided to investigate how these input quantities would impact the uncertainty of the measurand, and, based on the results, grouped them into those that should be incorporated into uncertainty evaluation and those that should not. As an example, figure 2(a) shows the pressure dependence of the voltage of an SOFC single cell that was measured while the atmospheric pressure was varied. The measurement was conducted in three different conditions: open circuit voltage (current: 0 A); the fuel utilization rate at 50 % (current: 24.4 A); and the fuel utilization rate at 70 % (current: 34.1 A). The slopes of the plots represent the impact of the atmospheric pressure on the uncertainty of voltage measurements when the current and gas flow rate are kept constant and the pressure deviates from the set value. The pressure dependence of the cell voltage obtained from the data shown in the graphs is approximately 0.3 V/Pa at around 1 atmospheric pressure, demonstrating that if the measurement is made within the normal range of atmosphere, the impact that the atmospheric pressure variation has on measurand V remains small. Incidentally, in our standard, the uncertainty of the measurand is to be kept at approximately 1 % or below as a rough target. SOFC varies not only by geometry and size, but also by the If the object has more than one temperature measurement points as is always the case with large cells or stacks, it was decided that manufacture has to provide the tolerance range of temperature distribution and the measurement has to be conducted within the range. It was also decided that the temperature distribution should not be included in the uncertainty evaluation. It was experimentally confirmed that the temperature distribution could become negligible using 100 cm2 or such a level cells. temperature distributionNot to be included in uncertainty evaluation, because it is not possible to change ambient pressure with common test equipment. It was alternatively decided to report atmospheric pressure. It may become problematic if atmospheric pressure differs much from 1 atm. Normally it was confirmed by experiment that this effect on the performance is small.atmospheric pressure±1 % of readingoxidant pressure±1 % of readingfuel pressurenot incorporated (compromized with the prescription that the test should be conducted under the condiiton that it does not affect the performance)n/an/an/an/an/a±1 % of readingnot prescribed particularly with the condition that it does not attect the performanceoxidant temperaturenot incorporated (compromized with the prescription that the test should be conducted under the condiiton that it does not affect the performance)n/a±1 % of readingnot prescribed particularly with the condition that it does not attect the performancefuel temperaturepartially achieved (e.g. presentation of the composition table published by gas manufacturer)depends on the method of mixture supply (cannot change indivisually)O2:±0.3 mole% (N2:balance)O2:±0.3 mole% (N2:balance)oxidant compositionpartially achieved (e.g. presentation of the composition table published by gas manufacturer)depends on the method of mixture supply (cannot change indivisually)H2, H2O, N2:±2 mole%、CO, CO2, CH4:±1 mole%H2, H2O, N2:±1 mole%、CO, CO2, CH4:±0.25 mole%fuel compositionincorporatedexperimental (rated flow rate±10~20 %)±1 % of reading±1 % of rated valueoxidant flow rateincorporatedexperimental (rated flow rate±10~20 %)±1 % of reading±1 % of rated valuefuel flow rateincorporatedexperimental (rated temperature±50 ℃)±1 % (℃) of rated temperature±1 % (℃) of set temperatureunit temperatureincorporatedexperimentally decided±1 % of rated current±1 % of rated current (in case of current control)currentincorporatedexperimentally decided±0.5 % of OCV±1 % of set value (in case of voltage control)voltageaction to unceratinty evaluationsensitivity coefficientinstrument uncertaintytolerable range of variationinput parameternot incorporatednot incorporatednot incorporatednot incorporated±3kPa (atmospheric pressure operation)±3kPa (atmospheric pressure operation)Table 1. Range of allowable variations and instrument uncertainty set by the draft standard and the feasibility of uncertainty evaluation of each input quantity

元のページ 

page 31

※このページを正しく表示するにはFlashPlayer10.2以上が必要です