The aim of this project was to investigate the sensitivity of the index test with focus on the behavior of the guide vane and runner blade angles. This was achieved by comparing the cam curves from an index test with a design of experiment. Index tests are a popular method for determining cam curves for Kaplan turbines. Index tests, as they are performed today, do not take into account a suspected hysteresis phenomenon in the guide vane and runner blade mechanism. If and how this affects the cam curve is today unknown which this project has investigated. An experiment was designed based on the “design of experiment” method. The difference from a traditional index testing is that the measurements were performed in a random order and only two propeller curves were plotted, but with eight replicates. Ten different guide vane angles were used; five for each runner blade angle and a total of 80 measurements were performed over the span of two days. Then, the measurements were compared with the results from an index test performed the day before on the same machine, a full-scale turbine in the Lule Älv-river using the same equipment. When analyzing the distribution of the efficiency, it has a larger spread on the left hand side of the propeller curve compared to the best efficiency point and negligible variation around the best efficiency point. The difference in the cam curves between the two methods was smaller than the estimated measurement error at the best efficiency point. The peak efficiency in the experiment was found for a slightly higher flow rate at one of the runner blade angles but this point was between two settings at the index test. The variations in efficiency are at their largest before the peak and may vary between a high efficiency and significantly lower efficiency at this point. Additional measurements may reduce the risk for choosing a guide vane angle at which such behavior occurs. A constant difference in the readings between the fixed scale and the station sensor were observed for the runner blade mechanism. This difference varied with the runner blade angle and amounted to 0.26° for 1° and 0.70° for 4° runner blade angle.For the guide vane mechanism no hysteresis could be validated. A difference between the fixed scale and the station sensor which varied depending on the direction was spotted and could be validated with 95% significance. The difference in the readings from the fixed scale and the station sensor was in average 0.15° larger when increasing the guide vane angle than when decreasing the angle. ; Validerat; 20150915 (global_studentproject_submitter)


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Uncertainties in Kaplan Cam Curve


    Beteiligte:

    Erscheinungsdatum :

    2015-01-01


    Anmerkungen:

    Local a1a8c2fa-e1b1-43a9-b041-501b280e8cf1


    Medientyp :

    Hochschulschrift


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Klassifikation :

    DDC:    600 / 629