Search for a command to run...
Abstract The objective of this paper is to present a suite of diagnostic methods and tools which have been developed to analyse and understand production performance degredation in wells lifted by ESPs in the Mangala field in Rajasthan, India. The Mangala field is one of the world’s largest full field polymer floods, currently injecting some 450kbbl/day of polymerized water, and a significant proportion of production is lifted with ESPs. With polymer breaking through to the producers, productivity and ESP performance in many wells have changed dramatically. We have observed rapidly reducing well productivity indexes (PI), changes to the pumps head/rate curve, increased inlet gas volume fraction (GVF) and reduction in the cooling efficiency of ESP motors from wellbore fluids. The main drivers for the work were to understand whether reduced well rates were a result of reduced PI or a degredation in the ESP pump curve, and whether these are purely down to polymer or combined with other factors, for example reduced reservoir pressure, increasing inlet gas, scale buildup, mechanical wear or pump recirculation. The methodology adopted for diagnosis was broken in 5 parts – 1) Real time ESP parameter alarm system, 2) Time lapse analysis of production tubing pressure drop, 3) Time lapse analysis of pump head de-rating factor, 4) Time lapse analysis of pump and VFD horse power 5) Dead head and multi choke test data. With this workflow we were able to break down our understanding of production loss into its constituent components, namely well productivitiy, pump head/rate loss or additional tubing pressure drop. It was also possible to further make a data driven asseesment as to the most likely mechanisms leading to ESP head loss (and therefore rate loss), to be further broken own into whether this was due to polymer plugging, mechanical wear, gas volume fraction (GVF) de-rating, partial broken shaft/locked diffusers or holes/recirculation. In some cases a specific mechanism was compounded with an associated impact. For example, in ESPs equipped with an inlet screen, heavy polymer deposition over the screen was resulting in large pressure drops across the screen leading to lower head, but this also resulted in higher GVFs into first few stages of the pump, even though the GVF outside the pump were low, leading to further head loss from gas de-rating of the head curve. With knowledge of the magnitude of production losses from each of the underlying mechanisms, targeted remediation could then be planned. The well and pump modelling adopted in the workflow utilise standard industry calculations, but the combination of these into highly integrated visual displays combined with time lapse analysis of operating performance, provide a unique solution not seen in commercial software we have screened. The paper also provides various real field examples of ESP performance deterioration, showing the impact of polymer deposition leading to increased pump hydraulic friction losses, pump mechanical failure and high motor winding temperature. Diagnoses based on the presented workflow have in many cases been verified by inspection reports on failed ESPs. Diagnosis on ESPs that have not failed cannot be definitive, though the results of remediation (eg pump flush) can help to firm up the probable cause.