Search

Geochemical analysis of 64 oils sampled from an Ordovician carbonate reservoir in the Tuoputai region was undertaken to study the composition of molecular markers. All the oils have similar geochemical characteristics and belong to a single oil family. They are presumed to derive from the same source kitchen and have similar oil charging history. A histogram of homogenization temperatures (T h) of aqueous inclusions in reservoir rocks shows a bimodal distribution pattern, indicating that the Ordovician reservoir has been charged twice. Coupling the measured T h (°C) with the burial and geothermal histories reconstructed using 1D basin modeling, we relate the homogenization temperature to the relevant geological ages: i.e., 425–412 and 9–4 Ma, corresponding to the Middle to Late Silurian and the Miocene to Pliocene, respectively. The oil filling orientation and pathways are traced using molecular indicators related to alkyldibenzothiophenes and benzo[b]naphthothiophenes. The oil charging orientation is from south to north generally. It can be predicted that the Ordovician reservoirs were sourced from a kitchen located to the south of the Tuoputai region, most probably between the Awati and Manjiaer Depressions. Traps located in the southern side of the Tuoputai region, along the oil charging pathways, should therefore be preferred oil exploration targets.

In order to increase the installed capacity, the refurbishment of Gondo high-head power plant required a modification of the existing surge tank by installing a throttle at its entrance. In a previous study, the geometry of this throttle was optimized by physical modeling to achieve the target loss coefficients as identified by a transient 1D numerical analysis. This study complements previous analyses by means of 3D numerical modeling using the commercial software ANSYS-CFX 19 R1. Results show that: (i) a 3D computational fluid dynamics (CFD) model predicts sufficiently accurate local head loss coefficients that agree closely with the findings of the physical model, (ii) in contrast to a standard surge tank, the presence of an internal gallery in the surge tank proved to be of insignificant effect on a surge tank equipped with a throttle, as the variations in the section of the tank cause negligible local losses compared to the ones induced by the throttle, (iii) CFD investigations of transient flow regimes revealed that the head loss coefficient of the throttle only varies for flow ratios below 20% of the total flow in the system, without significantly affecting the conclusions of the 1D transient analysis with respect to minimum and maximum water level in the surge tank as well as pressure peaks below the surge tank. This study highlights the importance of examining the characteristics of a hydraulic system from a holistic approach involving hybrid modeling (1D, 3D numerical and physical) backed by calibration as well as validation with in-situ measurements. This results in a more rapid and economic design of throttled surge tanks that makes full use of the advantages associated with each modeling strategy.

In order to increase the installed capacity, the refurbishment of Gondo high-head power plant required a modification of the existing surge tank by installing a throttle at its entrance. In a previous study, the geometry of this throttle was optimized by physical modeling to achieve the target loss coefficients as identified by a transient 1D numerical analysis. This study complements previous analyses through 3D numerical modeling using the commercial software ANSYS-CFX 19 R1. Results show that: (i) a 3D CFD model predicts sufficiently accurate local head loss coefficients that agree closely with the findings of the physical model; (ii) in contrast to a standard surge tank, the presence of a connecting gallery proved to be of insignificant effect on a surge tank equipped with a throttle, as the variations in the section of the tank cause negligible local losses compared to the ones induced by the throttle; (iii) CFD investigations of transient flow regimes revealed that the head loss coefficient of the throttle only varies for flow ratios below 20% of the total flow in the system, without significantly affecting the conclusions of the 1D transient analysis concerning minimum and maximum water level in the surge tank as well as pressure peaks below the surge tank. This study highlights the importance of examining the characteristics of a hydraulic system from a holistic approach involving hybrid modeling (1D, 3D numerical and physical) backed by calibration as well as validation with in-situ measurements. This results in a more comprehensive design of throttled surge tanks that makes full use of the advantages associated with each modeling strategy.