Early Stage of Commingled Water Injection in a Brownfield: Challenges and Lessons Learnt

Abstract

Water injection was implemented in a 30-year old brownfield offshore Sarawak, Malaysia in August 2016. Seawater is processed at a Water Injection Facility (WIF) and sent to four injectors, each injecting commingled into two or three different reservoirs. This paper discusses on challenges faced in initial start-up of water injection in a brownfield including the inability to meet target injection rate, frequent WIF trips and off-spec injection water, metering issues, as well as mitigation measures and lessons learned. Initially, the injectors were able to take in only 33% of target injection volume as per the FDP plan. To remedy this, a ramp-up injection procedure was introduced to allow the injectors to gradually take in more water until the target injection rate could be achieved. A leaner and practical water quality SOP was devised to reduce injector downtime, particularly for satellite platforms, while ensuring water quality is not compromised. Injection fall-off testing was performed on the injectors to investigate the root cause of the injectivity issue through manipulation of downhole ICV. Through this exercise, it was discovered that the injection meters were not properly calibrated. A combination of these methods proved successful in improving injection rate of the water injectors. Initial SOP developed for the injection water quality required testing of water quality at each sampling point including at unmanned satellite platforms, prior to recommencement of water injection post WIF shutdown. This is despite the duration of shutdown being shorter than the frequency of required sampling, which led to prolonged injection downtime. The requirement for water sampling for satellite platforms were modified to be less stringent while still maintaining good water quality. As a result, there was an improvement in WIF uptime from 92% in second month of injection to 99% in the fifth month. The fall-off testing provided valuable information in terms of well and reservoir data. Careful and specific operational steps were required to adjust the downhole ICVs during fall-off testing, as opposed to hard shut-in of the water injectors which would cause backpressure and tripping of the WIF. Adjustment of the surface-controlled ICVs allowed sequential testing of different zones, which successfully shortened the total testing duration by 25%. The fall-off test also revealed that an injector was injecting into a reservoir which did not benefit any producers, and that the flowmeters for certain injectors were not calibrated properly. Through these efforts, injection rates were successfully increased by 25 kbwpd, from 35% to 75% of the total injection target, within six months of its implementation. Water injection start-up challenges and mitigation methods are not often discussed in literature, such as adjustments needed to achieve target injection rate, operational steps in well testing for commingled injectors, and finding the optimum balance between quality and practicality of injected water testing. It is hoped that the issues and strategy in this field will serve as lessons learnt for upcoming water injection projects in this and nearby fields.