November 20, 2008
Advanced Search

Login:

Password:

Forgot your password?
Register now

Home / Issue Archive / 2007 / April #4 / Zapolyarnoye Oil and Gas-condensate Field Development Prompts Modification of Field Gas Treatment Units

№ 4 (April 2007)

Zapolyarnoye Oil and Gas-condensate Field Development Prompts Modification of Field Gas Treatment Units

The period of UKPG-3S operation at the Zapolyarnoye oil and gas condensate field before February 2005 saw the introduction of new field technologies and equipment. Design and construction faults were eliminated, and the technological process of gas dehydration, complicated by conditions of unstable equipment operation, was adjusted.

By Artyom Golubov

Share it!

During that period, the field faced problems inherent in the development of all large fields. Designed technical solutions turned out to be less than optimum due to specific features of the field and the equipment utilized.
Some of these features were considered and discussed but not incorporated into the project; others arose unexpectedly, which indicated an inadequate level of the scientific and methodological basis for the selected technical solutions.


The UKPG-3S gas complex treatment unit consists of two shops. Each shop is equipped with six processing trains with capacity of 10 mcm/day. The main technological gas treatment equipment of each train includes an inlet separator of "wash" type and a GP-1924 absorber. Gas is fed to the processing trains of the shops through individual pipelines from a horizontal reservoir 1,020 mm in diameter that merges flows of gathering lines from several well clusters. Gas is supplied to the reservoir from below and distributed to the processing trains through a side generatrix of the reservoir.


Fig. 1 shows averaged data for the liquid entrainment from the separator and absorber of five processing trains, Nos. 7, 8, 10, 11 and 12.
As can be seen from Fig. 1, there is irregularity in the entrainment from the units of different processing trains, with the trend of change of the entrainment value from one absorber to another replicating the trend of change of the entrainment from separators.


An analysis of separator function shows that both the efficiency of the units and their entrainment are inversely related to the liquid load on the unit, as both factors deteriorated with an increasing load. It is evident that the irregularity is a result of the reservoir design. The liquid entrainment from the separators is mainly composed of condensate (60-70 percent) and water-methanol solution (WMS). The entrainment from the 410-420 mcm/hour absorber is composed of condensate that at times includes a negligible presence of glycol.nto absorbers and condensate entrainment into gas pipelines negatively impact the function of the separation and regeneration assemblies. The experimental data indicate that the presence of hydrocarbon liquid reduces glycol gas dehydration efficiency and quality significantly (by 7-10 percent), as pressure decreases and absorption temperature increases.


According to observations of engineers at the field, a movement of liquid mass is periodically evident in this section of the reservoir.
Thus, the non-uniformity of the liquid load (mainly, condensate) on the apparatus of the processing trains is quite logically accounted for by hydrodynamic conditions in the reservoir.


To resolve the problem, the liquid level in the reservoir should be removed by installing a receiver of 2,000 mm diameter. The receiver must connect to the bottom of the reservoir through several pipes, with the liquid outlet equipped with an automatic level controller.


The data displayed in Fig. 1 prove extreme inefficiency of inlet separators' performance.


An evaluation of efficiency was carried out for the separation (recovery) of aqueous components of the liquid phase -  WMS in the separator, diethylene glycol (DEG) in the absorber, and condensate in two adjacent units. Using a special chamber, a visual examination of the aggregative state of the WMS-condensate and rich DEG-condensate mixtures was made under working pressure (8 MPa). The evaluation of separation efficiency showed that over 90 percent of WMS and about 60 percent of condensate were recovered in the separator. In the absorber, if the recovery of glycol approximates 100 percent, the recovery of condensate, at around 85 percent, is considerably higher than that in the separator. There is an essential design difference between the separator and the separation unit of the absorber, as the latter, in addition to vortex elements, has elements where the agents being separated do not mix and where exist conditions for coalescence of the agents on fabric and wire mesh surfaces.


The visual observation of the agents showed that mixing WMS and rich DEG with condensate resulted in formation of trice emulsions that resisted decomposition for hours.


No signs of foam formation above the emulsion surface were found in the course of the visual observation at 8 MPa of pressure. Based on observations, a conclusion may be drawn that the emulsions occurred during agitation and recirculation of liquids at the mass-transfer trays are the cause of the units' unusual behavior. The emulsions coalescence at the filtration elements results in considerably higher efficiency of the hydrocarbon component recovery in the absorber compared to that of the separator.


For years there have been debates about which field technology methods best protect against liquid produced with the use of inlet separators. The CKBN company has developed its concept using GP 1300.01 (Yamburg GCF) separators and GP-1924 (Zapolyarnoye OGCF) "wash" separators. The UKPG-3S units have four trays of centrifugal elements, with two separation trays (at the upper and lower positions) and two mass-transfer trays (at the middle positions) similar to ones used in absorbers. Fresh water, which is usually a reflux of glycol regeneration equipment, must be fed to these trays in contact with gas.


A "wash" separator dilutes the liquid entrained from the inlet separation tray with fresh water at two subsequent trays. Being depleted of salts in a degree corresponding to the ratio of washing fluid to the entrainment from the first tray, the liquid is fed to the upper separation tray, along with the entrainment from mass-transfer stages. The last separation stage partly recovers this low salt content solution, and the entrained liquid salts glycol to a lesser degree than an apparatus which would have only one separation stage.


It is not difficult to see that the concept of this apparatus provides for existence of an entrainment from all the trays. However, this entrainment is harmful not only because it contains salts but also because the dehydration efficiency decreases and the load on the glycol regenerator increases. The application of mass-transfer trays to feed fresh water into gas flow seems quite unreasonable, since it is possible to realize this much easier, without expensive GPR-340 elements, and there is no need to "wash" the gaseous phase, which does not contain salts.


There is another view that a similar protection against salts could be achieved by installation of several merely separation trays. In such a system, each next tray would have a higher efficiency due to less liquid load. Certainly, the separation elements of both trays must be specialized for different flow conditions concerning quantity and dispersity of the liquid phase. Taking into account the conditions at the Zapolyarnoye field, it would be most practical to use such a unit for the protection of the absorber against produced and condensation liquids. The unit design must not include vortex injection elements, as they are undoubtedly the main source of the trice emulsion formed during the multiple recirculation of the WMS-condensate mixture in high-gradient vortex flow.


For the purpose of minimizing the ingress of condensate into absorbers, the design of the GP-1924 "wash" separator is unacceptable. It would be expedient to renovate it for a higher efficiency.


Based on the operation analysis of the GP-778 (Yamburg field) and GP-1181 (Yamsovey field) apparatus, the prediction for a new GP-1924 apparatus functionality is as follows: it will function at design capacity with a consistently high glycol entrainment from the mass-transfer section to the separation section.


A complex study of the functionality of the apparatus was performed to evaluate the efficiency of the modified mass-transfer trays of the absorbers, as well as their ability to perform continuous effective operation. In a capacity range of 60-100 percent of the design value, the liquid entrainments to the filter trays were measured by a method developed by UNIPR - YGD, and the dew point temperature (DPT) was determined. It was noted that detection of condensation boundary on the condensate-wetted mirror is rather difficult, so the DPT determination for gas at the Zapolyarnoye OGCF is less reliable than that of the Yamburg field. This is likely to concern the reliability of measurements with "Kong-PRIMA" device.


Data processing for actually measured DPT showed that the averaged mass-transfer efficiency of the GP-1924 unit was equal to 1.1. For comparison, the mass-transfer efficiency of GP-1181 Yamsovey design version (before modernization) was approximately 1.5, according to our data. After modernization by CKBN at the Yubileynyi, the efficiency was 1.2-1.4; for GP-502 absorbers (in design version as well) efficiency was 1.6-1.7; and after modernization with Mellopak+Zultser and MKN packings it was 1.9-2.0.


The reasons for such low efficiency of the UKPG-1S units are the presence of hydrocarbon condensate and disturbance of backflow when glycol entrains from the mass-transfer section to the filtering section.


Due to a low enough temperature (0 to +3 C) in the initial period of field operation, such efficiency of the units is quite acceptable, and gas quality conformed to GOST requirements. However, as pressure decreases and temperature increases, the conditioned dehydration of gas will become problematic. Fig. 2 shows the difference in the gas dehydration efficiency at neqv = 1.1, 1.5, and 2.0 for operating parameters of 5 MPa pressure; +15 C temperature; 99.2 DEG solution; 3 cu. m/hour DEG solution feed; 420 mcm/hour gas flow rate; and 0.2 moisture content at the inlet.


Fig. 3 represents the dependence of the liquid entrainment from the mass-transfer section to the tray of cartridge filters for several type of apparatus. One should note that the GP-778 absorbers at the Yamburg field are equipped, just as  the GP-1181 and GP-1467 apparatus, with mass-transfer trays made up of GPR-340 vortex injection elements. Despite the considerably less internal volume of the mass-transfer section and fewer trays (four instead of six), the absorbers have an essentially greater value of critical capacity, at which the entrainment increases appreciably, compared to the GP-1181 and GP-1467 apparatus. This effect is completely accounted for by a larger quantity of the separation elements in the mass-transfer trays, namely 199 pieces in the GP-778 absorber against 164 in the GP-1181 and 182 in GP-1467.


The GP-1467 apparatus, which was installed at UKPG-1S in accordance with the design,  was made without taking into account the recently found effect of the trice emulsion. According to predictions, the new apparatus' performance parameters should have taken an intermediate position between curves for the GP-778 and GP-1181, as shown in Fig. 3. In practice, however, such evaluation of the apparatus' performance proved inaccurate. A decrease of the capacity to about half of the predicted value during tests had no results in removal of entraining to the cartridge filter, which is, apparently, a consequence of ingress of a large quantity of condensate forming a trice emulsion with glycol at the trays of the vortex recirculating elements.


Thus, it may be assumed that the main reason for ingress of condensate from the separator to the absorber is the use of the GPR-340 vortex injection elements (and probably the GPR-515 vortex separation elements as well) at both the apparatus.

Conclusions and Recommendations

1. The presence of a considerable amount of hydrocarbon condensate in Cenomanian gas of the Zapolyarnoye field and an underestimation of this factor in the design of the gas dehydration unit, as well as unsuccessful design solutions with respect to the inlet reservoir, separators and the dehydration absorber, are the factors that reduce potentialities of the field in terms of quality assurance and dehydration unit capacity.


2. The above-listed factors lead to a considerable deterioration in gas preparation quality and to an increase of equipment maintenance costs.


3. The inlet reservoir needs to be equipped with liquid receivers to prevent an occurrence of the level and to provide an equal distribution of liquid to processing trains.


4. Both the UPKG units should be redesigned, eliminating the use of vortex injection elements.


5. It is expedient to remove the gas washing function from the inlet "wash" separator and to transform it through reconstruction into a high-quality separator with several separation trays to best protect the glycol dehydration process in absorbers from liquid hydrocarbons.

Share it!
/
Copyright © 2008 Eurasia Press, Inc. (USA). All rights reserved.
Web programming by Iflexion
Copyright © 2008 Eurasia Press (www.eurasiapress.com)